Natural systems for wastewater treatment in warm climate regions

DISSERTATION Title NATURAL SYSTEMS FOR WASTEWATER TREATMENT IN WARM CLIMATE REGIONS Author LIVIA GARCIA GIL Tutor MARIANNA GARFÍ JOAN GARCÍA Department HYDRAULIC, MARITIME AND ENVIRONMENTAL ENGINEERING Intensification ENVIRONMENTAL ENGINEERING Date JULY 2014 Natural Systems for Wastewater Treatment in Warm Climate Regions Page 2 Natural Systems for Wastewater Treatment in Warm Climate Regions Page 3 ABSTRACT Water scarcity and the high costs of new water supply technologies are the two major factors responsible for the increasing recognition of the importance to conserve water resources by wastewater treatment, reuse or reclamation. Sustainability of sanitation systems should be related to low cost, low energy consumption and operation and maintenance requirements, especially for small communities in developing countries. Hence, natural systems for wastewater treatment seem to be a suitable solution. In this study, a review of two natural systems, constructed wetlands and stabilization ponds, was carried out. Thereby, strengths and weakness of both systems have been analysed. Furthermore, this dissertation evaluates the robustness of a pilot-scale hybrid constructed wetland system to cope with a heavy rain episode, a characteristic phenomenon in tropical climate regions. During three months (from June to September 2013), the pilot plant operated under an input flow of 33 l/hour (0.27 m/day HLR in vertical CWs). Under these conditions, the system showed good mass removal rates for all contaminants, (96.6% for TSS, 95.5% for BOD5, 77.6% for COD and 90.8% for NH4-N). These results were compared to that obtained in previous studies carried out during the coldest months of the year. In September 2013, a heavy rain episode was simulated. The pilot plant operated under an input flow of 330 l/hour (300l/h of potable water plus 33 l/h of real wastewater) for one hour. The removal rates were high (above 77.6% for all parameters), and the contaminants concentrations seemed to return to the normal values about 7 hours after the campaign. In summary, it can be concluded that the technology of constructed wetlands is a valid solution for wastewater treatment for small communities in warm climate regions. Likewise, these systems can cope with sharp fluctuations in flow to be treated. Keywords: BOD, COD, constructed wetlands, free water surface, horizontal subsurface flow, hydraulic load, natural systems, NH4-N, stabilization ponds, TSS, vertical subsurface flow, wastewater stabilization ponds, and wastewater treatment. Natural Systems for Wastewater Treatment in Warm Climate Regions Page 4 ACKNOWLEDGEMENTS This research has been funded by the European Commission for the financial support of the NaWaTech project (Grant Agreement N°: 308336). I would like to be thankful to all those that helped me, especially Carlos, my laboratory partner, and Javi. Also thanks to my tutors Anna and Joan for helping and guide me in this work. To Victor, for his patience and support Natural Systems for Wastewater Treatment in Warm Climate Regions Page 5 TABLE OF CONTENTS ABSTRACT ......................................................................................................................... 3 LIST OF FIGURES: .............................................................................................................. 8 LIST OF TABLES: ................................................................................................................ 9 1. GLOSSARY ............................................................................................................ 10 2. ACRONYMS .......................................................................................................... 12 3. INTRODUCTION ................................................................................................... 13 4. OBJECTIVES .......................................................................................................... 15 5. NATURAL SYSTEMS FOR WASTEWATER TREATMENT ......................................... 16 6. PRELIMINARY AND PRIMARY TREATMENT ......................................................... 17 6.1. SCREENING .......................................................................................................... 17 6.2. UPFLOW ANAEROBIC SLUDGE BLANKET REACTOR ............................................. 17 7. CONSTRUCTED WETLANDS .................................................................................. 19 7.1. HISTORY AND INTRODUCTION: ........................................................................... 19 7.2. DESIGNS OF CONSTRUCTED WETLANDS: ............................................................ 21 7.2.1. Free Water Surface System (FWS) ....................................................................... 22 7.2.2. Subsurface Flow System (SSF) ............................................................................. 23 7.2.2.1. Horizontal Subsurface Flow System ............................................................. 23 7.2.2.2. Vertical Subsurface Flow System ................................................................. 25 7.2.3. Hybrid Systems .................................................................................................... 26 7.3. HYDROLOGICAL PARAMETERS ............................................................................ 26 7.4. REMOVAL MECHANISMS IN CONSTRUCTED WETLANDS: ................................... 28 7.4.1. Mechanisms of Suspended Solids ....................................................................... 29 7.4.1.1. Suspended Solids in Free Water Surface Wetlands ..................................... 29 7.4.1.2. Suspended Solids in Subsurface Flow Wetland ........................................... 31 7.4.2. Mechanisms for Organic Matter ......................................................................... 32 7.4.2.1. Organic Matter in Free Water Surface Wetlands ........................................ 32 7.4.2.2. Organic Matter in Subsurface Flow Wetlands ............................................. 34 7.4.3. Mechanisms for Nitrogen .................................................................................... 35 7.4.3.1. Nitrogen in Free Water Surface ................................................................... 38 Natural Systems for Wastewater Treatment in Warm Climate Regions Page 6 7.4.3.2. Nitrogen in Subsurface Flow Wetlands ........................................................ 38 7.4.4. Mechanisms for Phosphorus ............................................................................... 39 7.4.4.1. Phosphorus in Free Water Surface .............................................................. 39 7.4.4.2. Phosphorus in Subsurface Flow Wetlands ................................................... 40 7.4.5. Mechanisms for Pathogens ................................................................................. 41 7.4.6. Mechanisms for Metals ....................................................................................... 43 7.4.7. Mechanisms for Other Contaminants ................................................................. 44 7.5. POTENTIAL HAZARD FOR MOSQUITO DEVELOPMENT ....................................... 46 8. WASTE STABILIZATION PONDS ............................................................................ 48 8.1. HISTORY AND INTRODUCTION ............................................................................ 48 8.2. CLIMATE, PHYSICAL AND BIOLOGICAL FACTORS: ............................................... 50 8.3. DESIGNS OF STABILIZATION PONDS: ................................................................... 51 8.3.1. Anaerobic ponds (AnP) ........................................................................................ 52 8.3.2. Facultative ponds (FP) ......................................................................................... 53 8.3.3. Maturation ponds (MP) ....................................................................................... 56 8.4. REMOVAL MECHANISMS IN STABILIZATION PONDS: ......................................... 57 8.4.1. Mechanisms for Suspended Solids ...................................................................... 57 8.4.2. Mechanisms for Organic Matter ......................................................................... 58 8.4.3. Mechanisms for Nitrogen .................................................................................... 59 8.4.4. Mechanisms for Phosphorus ............................................................................... 61 8.4.5. Mechanisms for Pathogens ................................................................................. 62 8.4.6. Mechanisms for Heavy Metals: ........................................................................... 64 8.5. ALGAE CONTROL: ................................................................................................. 65 8.6. ODOUR RELEASE AND CONTROL: ........................................................................ 66 8.7. UASBs VS. ANAEROBIC PONDS: ........................................................................... 67 9. WATER REUSE: ..................................................................................................... 68 10. CASE STUDY: Influence of heavy rain episodes on removal efficiency in three stage hybrid treatment wetlands ................................................................................... 69 10.1. DESCRIPTION OF THE PILOTE-SCALE TREATMENT WORKS: ................................ 69 10.2. DESCRIPTION OF THE TREATMENT PROCESS: ..................................................... 70 Natural Systems for Wastewater Treatment in Warm Climate Regions Page 7 10.3. HYDROLOGICAL PARAMETERS: ........................................................................... 72 10.3.1. Under design conditions: ............................................................................. 72 10.3.2. Under heavy rain conditions: ....................................................................... 73 10.4. SAMPLING STRATEGY: ......................................................................................... 73 10.4.1. Analytical methods: ...................................................................................... 74 10.4.2. Data treatment: ............................................................................................ 75 10.5. RESULTS: .............................................................................................................. 75 10.5.1. Performance of the treatment system under normal conditions: .............. 75 10.5.1.1. Chemical Oxygen Demand ........................................................................... 75 10.5.1.2. Biochemical Oxygen Demand ...................................................................... 76 10.5.1.3. Total Suspended Solids ................................................................................ 77 10.5.1.4. Ammonia-Nitrogen ....................................................................................... 78 10.5.1.5. pH, Eh and DO .............................................................................................. 79 10.5.2. Under heavy rain conditions: ....................................................................... 80 10.5.2.1. Chemical Oxygen Demand ........................................................................... 80 10.5.2.2. Total Suspended Solids ................................................................................ 81 10.5.2.3. Ammonia-Nitrogen ....................................................................................... 82 10.6. DISCUSSION: ........................................................................................................ 83 10.6.1. Discussion of the results under normal conditions: .................................... 83 10.6.2. Discussion of the results under heavy rain conditions: ............................... 84 11. CONCLUSIONS AND RECOMMENDATIONS: ........................................................ 86 REFERENCES: ................................................................................................................... 88 Natural Systems for Wastewater Treatment in Warm Climate Regions Page 8 LIST OF FIGURES: Figure 1: Schematic diagram of an UASB (Van Haandel and Lettinga, 1994) ................ 18 Figure 2: Cumulative number of WW treatment plants based in CWs over the last years in various European regions (Puigagut et al., 2007) ...................................................... 20 Figure 3: Treatment wetland types (Kadlec and Wallace, 2008) ................................... 21 Figure 4: Schematic drawing of common reed, Phragmites communis (Garcia and Corzo, 2008) .................................................................................................................... 22 Figure 5: Cross section of a FWS CW (Garcia and Corzo, 2008) ..................................... 22 Figure 6: Profile of a three-zone FWS CW cell (U.S. EPA, 2000) ..................................... 23 Figure 7: Cross section of a Horizontal SSF system (Garcia and Corzo, 2008 adapted) . 23 Figure 8: Cross section of a Vertical SSF system (Garcia and Corzo, 2008 adapted) ..... 25 Figure 9: An hybrid wetlands system: vertical SSF followed by an horizontal SSF (Kadlec and Wallace, 2008) ......................................................................................................... 26 Figure 10: Removal mechanisms that dominate in FWS CWs (U.S. EPA, 2000) ............ 29 Figure 11: Processes affecting particulate matter removal and generation in FWS wetlands (Kadlec and Wallace, 2008)............................................................................. 30 Figure 12: Simplified scheme of organic matter removal processes (Garcia and Corzo, 2008) ............................................................................................................................... 34 Figure 13: Common nitrogen species present in WW ................................................... 35 Figure 14: Phosphorus removal rate constants for the Orlando Easterly Wetlands, Florida (Kadlec and Wallace, 2008) ................................................................................ 40 Figure 15: Typical pond layouts systems (adapted from Gloyna, 1971) ........................ 51 Figure 16: Typical cross-section of a FP (Gloyna, 1971) ................................................. 53 Figure 17: Symbiosis of algae and bacteria in FP and MP (Mara, 1976) ........................ 54 Figure 18: Basic biological interacions in a FP with emphasis on solids and organics transformations (Shilton, A., 2005) ................................................................................ 59 Figure 19: Top view of the pilot-scale treatment works (adapted from Donoso, 2013) 69 Figure 21: Studied pilot-scale treatment plant (Donoso, 2013) .................................... 70 Figure 22: Cross section of the vertical SSF CW ............................................................. 71 Figure 23: Cross section of the horizontal SSF CW ......................................................... 71 Figure 24: Cross section of the FWS CW ........................................................................ 72 Figure 20: Hybrid CW System (Imhoff tank was remplaced by a HUSB reactor) (Avila et al., 2013) ......................................................................................................................... 72 Figure 25: Plan diagram of the pilot plant with the corresponding sampling points (Imhoff was replaced by a HUSB reactor) (Avila et al., 2013) ........................................ 74 Figure 26: Average (± s.d) values of COD in the effluent of the different stages of treatment ........................................................................................................................ 76 Natural Systems for Wastewater Treatment in Warm Climate Regions Page 9 Figure 27: Average (± s.d) values of BOD5 in the effluent of the different stages of treatment ........................................................................................................................ 76 Figure 28: Comparison of the average values between COD and BOD5 in each stage of treatment ........................................................................................................................ 77 Figure 29: Average (± s.d) values of TSS in the effluent of the different stages of treatment ........................................................................................................................ 78 Figure 30: Average (± s.d) values of NH4-N in the effluent of the different stages of treatment ........................................................................................................................ 78 Figure 31: NH4-N concentration in cold and warm season (adapted from Amigó, 2013) ........................................................................................................................................ 79 Figure 32: Evolution of the COD concentration in each stage of the system during the heavy rain campaign ....................................................................................................... 81 Figure 33: Evolution of the TSS concentration in each stage of the system during the heavy rain campaign ....................................................................................................... 82 Figure 34: Evolution of the NH4-N concentration in each stage of the system during the heavy rain campaign ....................................................................................................... 83 LIST OF TABLES: Table 1: Typical construction and O&M costs for natural systems (W.P.C.F, 1990) ...... 16 Table 2: Main advantages of the different CWs designs (W.P.C.F., 1990) ..................... 22 Table 3: Removal Mechanisms in CWs (Moshiri, 1993) ................................................. 28 Table 4: Guidelines for Metal Concentrations in WW (U.S. EPA, 2002a)....................... 43 Table 5: Factors that have been proposed to cause or influence disinfection in WSPs (Shilton, 2005) ................................................................................................................ 62 Table 6: Features of the three main mechanisms of sunlight disinfection (adapted from Shilton, 2005) .................................................................................................................. 63 Table 7: Removal rates of contaminants (%) at the different types of WSPs (CENTA, 2008) ............................................................................................................................... 65 Table 8: Sampling strategy ............................................................................................. 74 Table 9: Average values (± s.d) of pH, Eh and DO in the effluent of the different stages of treatment ................................................................................................................... 79 Natural Systems for Wastewater Treatment in Warm Climate Regions Page 16 5. NATURAL SYSTEMS FOR WASTEWATER TREATMENT Natural systems for WW treatment are a biological system in which purification is carried out without energy input, and therefore reactions responsible for water purification occur very slowly (Salas et al., 2007). For this reason, hydraulic retention time in these systems can be even 100 times higher than in conventional treatment systems. This is the reason why they required larger land areas to treat the same water flow than in conventional systems. The main features of natural treatments are: - Reliability: natural systems are very reliable in extreme operating conditions. They can treat a variety range of WW and they work under a wide range of weather conditions (W.P.C.F., 1990) - Environmental benefits: aesthetic and wildlife are insured - Simplicity of the plants design: maintenance can be carried out by low skilled workers - Low operation and maintenance costs Natural WW systems are simple, cost-effective and efficient methods to treat WW. They are usually applied as secondary or tertiary treatment, allowing the removal of most of the bacteria, microorganism and organic matter. In tropical and subtropical developing countries where capital is scare but labour plentiful and relatively cheap, labour intensive schemes are economically and socially more advantageous (Mara, 1976). Hence, natural systems to treat WW are a feasible solution. Moreover in these regions, generally sufficient land is normally available. Nevertheless, few studies assessed the robustness of these systems during heavy rain episodes. Concept Capital Cost ($/m3·d) O&M Cost ($/m3·d) Slow rate infiltration1 800-2000 0.10-0.20 Rapid infiltration2 450-900 0.05-0.10 Overland flow3 600-1000 0.08-0.15 CWs4 500-1000 0.03-0.09 WSPs5 500-1000 0.07-0.13 Table 1: Typical construction and O&M costs for natural systems (W.P.C.F, 1990) 1 Includes an allowance for pre-treatment and storage, flow rate ≈ 400m3/d 2 With pre-treatment to primary, flow rate ≈ 400m3/d 3 Pre-treatment: screening or settling, flow rate ≈ 400m3/d 4 FWS type, pre-treatment: screening or settling, flow rate ≈ 400m3/d 5 No pre-treatment, flow rate ≈ 400m3/d Natural Systems for Wastewater Treatment in Warm Climate Regions Page 17 6. PRELIMINARY AND PRIMARY TREATMENT 6.1. SCREENING The first stage of WW treatment is the removal of large floating objects and heavy mineral particles. These materials can damage the equipment used, clog the pipes and accumulate on the surface of the secondary treatment systems. Coarse solids are usually removed by screens. The spacing between bars is usually 1525 mm. At small works, screens are raked by hand and, in order to facilitate this, the screens are inclined, commonly at 60o to the horizontal. 6.2. UPFLOW ANAEROBIC SLUDGE BLANKET REACTOR A primary treatment is strongly recommendable in order to reduce the organic loading and suspended solids of the effluent. Recently, anaerobic reactors have been becoming popular as primary treatment, including Upflow Anaerobic Reactors, Imhoff tanks and Hydrolytic Upflow Sludge Blanket reactors. UASBs are high-rate anaerobic digesters. They were developed in the 1970’s by Professor Lettinga, and they have been extensively tested at full-scale in tropical and subtropical regions, particularly in Brazil, Colombia and India (Mara, 2003). They have been used for the primary treatment of domestic and mixed WW and high-strength biodegradable industrial and agro-industrial WW. UASBs are reinforced-concrete structures with a short hydraulic retention time, of the order of 6-12 hours. As shown in figure 1, the raw WW is distributed across the base of the reactor and flows upwards through the sludge layer, what ensures the thorough contact between the WW and the anaerobic bacteria in the sludge. The liquor rises through the reactor, and during this time, the biodegradation of organic matter occurs, and reaches the “phase separator”. This is the important characteristic of this type of anaerobic reactor: it divides the reactor into its two constituent zones, the lower digestion zone and the upper settling zone. As the liquor rises through the settling zone, its Upflow velocity decreases due to the inclined surface of the phase separator, and the suspended sludge particles settle out. Finally the weight of the accumulated sludge particles exceeds the frictional force that keeps them on the inclined surfaces, and the settle down to the sludge layer. The phase separator enables an effluent with a very low suspended solids concentration to be discharged from the reactor. Natural Systems for Wastewater Treatment in Warm Climate Regions Page 18 Biogas bubbles are collected under the phase separator, from where the gas is easily extracted and can be re-used. Deflectors are placed between the phase separators units to prevent any biogas bubbles entering the settling zone where they would hinder sedimentation. Figure 1: Schematic diagram of an UASB (Van Haandel and Lettinga, 1994) UASBs produce quite large amounts of waste sludge, 0.2kg of sludge/kg of BOD removed. This is much less than in conventional activated sludge plants, but much more than in anaerobic ponds. In warm climates, UASB waste sludge can be simply dewatered on drying beds. Hydrolytic upflow sludge blankets (HUSB) are essentially UASB reactors operated at a lower HRT, from 2 to 5 hours, in order to avoid methanogenesis reaction wherever possible. In general, solids retention time in HUSB reactors is maintained for over 15 days in order to achieve high hydrolysis rates of WW solids. HUSB reactors have been recently investigated as a suitable primary treatment for CWs, mainly because they can provide effluents with lower TSS and COD concentration than standard primary treatments (Pedescoll et al., 2011). Natural Systems for Wastewater Treatment in Warm Climate Regions Page 19 7. CONSTRUCTED WETLANDS 7.1. HISTORY AND INTRODUCTION: Wetlands are land areas that are wet during part or all of the year. They are frequently transitional between uplands and continuously or deeply flooded systems (Kadlec and Wallace, 2008). In many coastal plain areas of the southeastern U.S. and in the poorly drained marshes and fens of the north, natural wetlands have historically been used as convenient receiving waters for WW discharges. Several of them, including the Houghton Lake fen in Michigan or the Florida cypress dones designed for WW management, were extensively studied in the U.S. and it has been recognised that their treatment capacity is quite unknown due to their variability and changes over time (W.P.C.F., 1990). The first attempts to use the wetland vegetation to remove various pollutants from water were in early 1950s. The first full-scale FWS was built in The Netherlands to treat WW from a camping site during the period 1967-1969. In late 1980s, soil was replaced with coarse materials (washed gravel) and this set-up has been successfully used since then (Vyzamal, 2005). Through the 1980s, a more thorough understanding has developed the specific strengths and weakness of CWs as treatment systems. This process has mirrored the increasing acceptance of the use of upland systems for WW renovation (W.P.C.F., 1990) and this treatment technology rapidly spread around the world. In 1990s, the increased needed for nitrogen removal from WW led to more frequent use of vertical flow CWs which provide higher degree of oxygenation in the bed, and the consequent removal of ammonia via nitrification. Few years later, in order to produce simultaneously nitrification and denitrification to remove total nitrogen lead to the use of hybrid systems (Vyzamal, 2005). The success of these new systems is probably due to the change of the last years towards a sustainable development and the more concern about the resource exploitation. For thirty years, WW treatment plants based in CWs have found greatest popularity in some areas (central and northern Europe) to treat WW of small communities, as shown in Figure 2. Nowadays, this system is used all around the world, including northern countries as well as southern countries (Garcia and Corzo, 2008). In developing countries, CWs are an attractive alternative to conventional WW treatment technologies. Natural Systems for Wastewater Treatment in Warm Climate Regions Page 20 Figure 2: Cumulative number of WW treatment plants based in CWs over the last years in various European regions (Puigagut et al., 2007) Wetlands have properties that make them unique among the major ecosystems groups. Ample water is important for most forms of biological productivity, and wetlands plants are adapted to take advantage of this abundant supply. Because of this, wetlands are among the most biological productive ecosystems on the planet (Kadlec and Wallace, 2008). Thereby, as they have a higher rate of biological activity than most ecosystems, they can transform many of the common pollutants that occur in conventional WW (Kadlec and Wallace, 2008). CWs are designed to take advantage of many processes that occur on natural wetlands but in a more controlled environment (Vymazal, 2005). They may be used to treat municipal WW, domestic, animal, mine water and industrial WW, as well as leachate and runoff. CWs may have several advantages compared to conventional and advanced WW treatment systems. Some of these advantages are:  Low cost of construction and maintenance.  Low energy requirements.  “Low-technology” system, it can be run by relatively low-skilled personnel.  Flexible systems and less susceptible to variations in loading rate than conventional treatment systems. The main disadvantage of CWs is the increased land area required, compared to conventional systems and their possible decreased performance during winter in temperate regions (Moshiri, 1993). Thus, CWs are a viable and advisable solution for tropical developing countries. Natural Systems for Wastewater Treatment in Warm Climate Regions Page 21 7.2. DESIGNS OF CONSTRUCTED WETLANDS: CWs can be classified based on the water flow pattern. CWs are classified in FWS systems with shallow water depths and SSF system with water flowing literally through the sand or gravel (U.S. EPA, 1988). There are subcategories under subsurface flow category which depend on the water flow, horizontal or vertical. Figure 3: Treatment wetland types (Kadlec and Wallace, 2008) There are many general functions of vegetation in wetlands, especially with regards to chemical processing and removal. This vegetation may be categorized by their growth habit with respect to the wetlands water surface as (Kadlec and Wallace, 2008): - Emergent soft tissue plants - Emergent woody plants - Submerged aquatic plants - Floating plants - Floating mats Emergent soft tissue macrophytes are the dominant life in wetlands and marshes, growing within a water-table ranging from 50 cm below the soil surface to a water depth of 150 cm or more (Moshiri, 1993). In general, these macrophytes have an extensive root and rhizome system, and the depth penetration of the root system as well as the sediment volume is different for different species. Some of the most common plant species used in CW design includes cattails (Typha), bulrushes (Scirpus) and common reed (Phragmites communis) (W.P.C.F., 1990). Oxygen is transported through the gas spaces to the roots and rhizomes by diffusion and/or by convective flow of air. Part of the oxygen may leak from the root systems into the surrounding rhizosphere, creating oxidized conditions Natural Systems for Wastewater Treatment in Warm Climate Regions Page 22 in the otherwise anoxic sediment and stimulating both decomposition of organic matter and growth of nitrifying bacteria (Moshiri, 1993). Figure 4: Schematic drawing of common reed, Phragmites communis (Garcia and Corzo, 2008) The main advantages of the two main designs are listed below. FWS Systems SSF Systems  Lower installation cost  Simpler hydraulics  Greater cold tolerance  Minimization of the insect vectors and odour problems  Greater assimilation potential per unit of land area Table 2: Main advantages of the different CWs designs (W.P.C.F., 1990) 7.2.1. Free Water Surface System (FWS) FWS systems can be understood as a modification of natural lagoons with a water depth of 0.3 and 0.4 meters, and plants (Garcia and Corzo, 2008). WW is directly exposed to the atmosphere, and water flows mainly through the leaves and stems of the plants. The selection of plant species type does not appear to be overly critical to assimilation capacity. This is because of the major role in assimilation played by the microbes that are attached to the plants and present in the wetland surface (W.P.C.F., 1990). The percentage of plant cover appears to be more important than the actual species composition but at the same time, this can suppose a problem with mosquito populations. Figure 5: Cross section of a FWS CW (Garcia and Corzo, 2008) Generally, the distribution of plants in a FWS wetland is not homogeneous, there may be open water zones with submerged plants and other zone fully vegetated. This Rhizomes Roots Outlet flow Inlet flow Natural Systems for Wastewater Treatment in Warm Climate Regions Page 23 heteroigenity may mean in practice different propierties and removal mechanisms in each zone. Figure 6: Profile of a three-zone FWS CW cell (U.S. EPA, 2000) The most common application for FWS wetlands is for advanced treatment of effluent from secondary or tertiary treatment processes (Kadlec and Wallace, 2008). 7.2.2. Subsurface Flow System (SSF) In SSF wetlands, water flows through the soil medium in contact with the roots and rhizomes of the plants. The water depth is from 0.3 to 0.9 meters. The biofilm grows adhered to the soil medium and the roots and rhizomes, and it is fundamental to the pollutants removal (Garcia and Corzo, 2008). In SSF systems, bulrush and common reed have the best properties for use due to their root development and sediment aeration potential (W.P.C.F., 1990). The main advantages of SSFs over FWSs are: bigger treatment capacity (they accept higher OLRs), lower risk of contact between WW and persons, and lower risk of mosquitoes appearance. Nevertheless, they are less useful for environmental restoration projects due to the lack of accessible water sheet (Garcia and Corzo, 2008). 7.2.2.1. Horizontal Subsurface Flow System In these systems, water flows horizontally through the granular bed. The water layer is 0.3-0.9 meters deep, and it flows 0.05-0.1 meters under the soil surface. They are characterized by working permanently flooded and with OLRs of 6g DBO/m2·day (Garcia and Corzo, 2008). Figure 7: Cross section of a Horizontal SSF system (Garcia and Corzo, 2008 adapted) (5) Outlet structure (3) Granular bed (1) Inlet structure (2) Impermeable layer structure (4) Emergent vegetation Natural Systems for Wastewater Treatment in Warm Climate Regions Page 24 - Waterproofing: it is necessary to underlay the bed by an impermeable membrane to prevent seepage and groundwater contamination. Depending on the local conditions, it can be enough to compact the soil, in other cases, it will be necessary a clay layer or a synthetic liner (U.S. EPA, 1988). - Inlet and outlet structures: CWs are systems that require a good distribution and collection of the water in order to achieve the estimated outputs. Thus, inlet and outlet structures have to be carefully designed and constructed. The WW from the primary treatment has to be collected into a sink, and then water will be homogenously distributed to the wetland. The effluent is collected at a perforated pipe which is at the bottom of the wetland and is connected to an inverse “L”-shaped pipe with adjustable level for water level control in the wetland (Garcia and Corzo, 2008). Outlet structure controls must be able to control depth of water in the wetlands especially for winter ice conditions where deeper wetland conditions are required to maintain treatment levels (U.S. EPA, 1988). - Granular bed: the inlet and outlet are often filled with coarse gravel in order to distinguish these structures from the granular bed. The bed must be clean (exempt from fines), homogeneous, hard, durable and able to keep its shape in long-term. Moreover, it has to support the growth of the emergent vegetation and biofilm. Diameters of 5-8 mm achieve good results. The performance of the system depends also on some hydraulic parameters, such as the hydraulic conductivity which determine the stream that can flow through the soil (Garcia and Corzo, 2008). - Vegetation: the most used specie is common reed Phragmites australis (Moshiri, 1993). The vegetation has five important functions in the process:  Increase and stabilize the hydraulic conductivity of the soil (Moshiri, 1993).  Supply oxygen to the heterotrophic microorganisms in the rhizosphere: around the roots, there are aerobic microenvironments where microbial processes take place, such as nitrification and aerobic removal of organic matter (Moshiri, 1993).  Roots and rhizomes create a suitable surface for the growth of the biofilm (Garcia and Corzo, 2008).  Temperature variation dampening: when plants grow, they reduce the light intensity incident on the soil surface, avoiding temperature gradient that may affect some processes. Moreover, vegetation protects from freezing (Garcia and Corzo, 2008).  Uptake of nutrients: modest contribution to the nutrients removal in urban WW, but it is higher in diluted WW (Garcia and Corzo, 2008). Natural Systems for Wastewater Treatment in Warm Climate Regions Page 25 7.2.2.2. Vertical Subsurface Flow System In these systems, water flows pulsed and vertically through the granular bed. Hence, the granular medium is not permanently flooded. The water layer is 0.5-0.8 meters deep, and they work with OLRs of 20g DBO/m2·day (Garcia and Corzo, 2008). The vertical wetlands have a higher treatment capacity than the horizontals (they require less treatment surface to treat the same OLR), but they are more liable of clogging (Garcia and Corzo, 2008). During the loading period, air is forced out of the soil; during drying period, atmospheric air is drawn into the pore-spaces of the soil and diffusive oxygen transport is enhanced, thus increasing soil oxygenation. This operational regime provides alternating oxidizing and reducing conditions in the substrate, stimulating nitrification-denitrification and phosphorus adsorption (Moshiri, 1993). Figure 8: Cross section of a Vertical SSF system (Garcia and Corzo, 2008 adapted) Vertical wetlands usually include aeration pipes. The aspects of waterproofing and vegetation are exactly the same as in horizontal wetlands. - Inlet and outlet structures: water is distributed through a piping net set out over the surface. Owing to the discontinuous flow, on cold climates, the piping net is buried 0.05-0.1 meters under the surface in order to prevent from freezing. The effluent is collected at a perforated piping net which is at the bottom of the wetland (Garcia and Corzo, 2008). - Granular bed: The bed must be heterogeneous, consisting of three horizontal layers with different gradation, which has to increase with the depth (coarse sand, gravel, and coarse gravel at the bottom) in order to prevent from a low/high flow speed (Garcia and Corzo, 2008). - Aeration pipes: these elements are used to air the bottom of the bed to enhance the aerobic degradation processes and the nitrification. In general, it is recommended the installation of 1 aeration pipe every 4 m2 (Kadlec et al, 2000). (4) Emergent vegetation (5) Outlet structure (1) Inlet structure Aeration pipes (2) Impermeable layer (3) Granular bed Natural Systems for Wastewater Treatment in Warm Climate Regions Page 32 TSS retention takes place on the firsts centimetres of the bed; and it concentration decreases exponentially along the depth (Garcia and Corzo, 2008). Reaction chemistry as noted previously for FWS wetlands can also occur in horizontal SSF wetlands. One use of horizontal wetlands has be sulphate-reducing systems to induce the precipitation of coppers, nickel and other metals (Egger, 1992). The performance of TSS removal is consistently high, around 90% producing outlet effluent with HLRs below 20mg/L (Garcia and Corzo, 2008). Removal efficiency for TSS is also closely related to input concentration, with lower efficiencies measured at low input concentrations. The critical HRT for achieving TSS removal efficiencies above 70% appears to be about 5 days (W.P.C.F., 1990). As mentioned above, most organic matter is removed in the inlet zone. This is the zone of the heaviest biosolids accumulation, where the greatest reductions in hydraulic conductivity occur. This zone can be termed the biosolids clogging distance. Clogging of the filter media is matter of concern because the bed may end up functioning, especially with high TSS loading (>50 mg/L) (Garcia and Corzo, 2008). 7.4.2. Mechanisms for Organic Matter One of the major constituents of raw and treated WW is organic matter (W.P.C.F., 1990), its removal is quite complex because is the result of the interaction of several physics, chemicals and biological mechanisms that occur simultaneously. The diverse array of sources of organic matter make characterization difficult, the total organic carbon and volatile solids (VS) measure the total amount of organic matter, the chemically oxidizable organic matter is measured as COD, and the biodegradable organic matter is determined by the BOD (U.S. EPA, 2000). Most regulatory agencies establish WW discharge permit limits based on BOD5 values (W.P.C.F., 1990). 7.4.2.1. Organic Matter in Free Water Surface Wetlands The mechanisms that regulate dissolved organic matter removal in wetlands include biodegradation, sorption and photolysis (U.S. EPA, 2000). The end products will depend on the presence or absence of oxygen. Areas of the wetland populated with dense emergent macrophytes can supply only a small fraction of the needed oxygen and most of the water column is anoxic, even though small microsites containing oxygen may be found adjacent to active plant roots. On the other hand, open areas of wetlands, containing submerged plants, have aerobic conditions throughout the wetland depth. Natural Systems for Wastewater Treatment in Warm Climate Regions Page 33 The effluent form the primary treatment contains some particulate organic matter as well as dissolved and colloidal fraction. This influent particulate organic matter may be entrapped within biofilm attached to emergent plant surface or accumulated on the wetland floor. In addition, the organic matter deriving from dead plant may accumulate also on the floor of the wetlands. The separation of the particulate organic matter would occur by the same mechanisms as those described for TSS (U.S. EPA, 2000). The soluble organic matter is removed by a number of separation processes, such as adsorption and absorption. This soluble material is more likely sorbed onto plant surface biofilm, and may be metabolized by organisms associated to this biofilm. The metabolic pathway and the end products of this metabolism will depend on the presence or absence of oxygen. The degree of sorption and its rate depend on the characteristics of the organic and the solid. Volatilization may also account for the loss of certain organics. However, organic matter entering a wetland receiving primary treatment will not contain significant quantities of VSs (U.S. EPA, 2000). Biological processes change the concentration and composition of organic matter in wetlands; these reactions include oxidation/reduction processes, hydrolysis and photolysis (U.S. EPA, 2000). Organisms will consume organic matter (and inorganic matter) to sustain life and to reproduce. The organic matter in WW serves as an energy source. Aerobic metabolism is the most efficient conversion of biodegradable materials to mineralized end products, gases, and biomass. Anoxic reactions use nitrates, carbonates, or sulphates as terminal electron acceptor, producing end products such as nitrogen oxides, free nitrogen, sulphur, etc. These reactions are typically less efficient than aerobic reactions and will not result in the reduction in BOD unless hydrogen or methane is produced (U.S. EPA, 2000). Much of the particulate organic matter will be hydrolysed, producing lower molecular weight organic compounds that are more soluble in water. In the presence of oxygen, these compounds will be oxidized by microbes to CO2, oxidized forms of nitrogen and sulphur and water. Under anaerobic conditions, these compounds will be converted to low molecular weight organic acids and alcohols. Under strict anaerobic conditions, methanogenesis will occur whereby these compounds are converted to gaseous end products of CH4, CO2 and H2. In the presence of sulphates, sulphur-reducing microbes will convert these low weight organic compounds to CO2 and sulphides. It has been observed that the removal efficiency depends on the input concentration. The low efficiency at low input concentration appears to be related to the internal Natural Systems for Wastewater Treatment in Warm Climate Regions Page 34 production of BOD5 in wetlands and possibly to insufficient substrate for microbes at these low concentrations (W.P.C.F., 1990). The rates of degradation are temperature dependent. Thus sediment organic matter may accumulate during the colder months and be more rapidly degraded in the spring and summer. 7.4.2.2. Organic Matter in Subsurface Flow Wetlands In horizontal SSFs, oxygen sources will be limited to some small amount of surface aerobic and plant-mediated transport, the predominant biological mechanism is likely to be anaerobic, while in vertical wetlands it seems to be more important the aerobic degradation of organic matter (Garcia and Corzo, 2008). Particulate organic matter is retained by filtration near the inlet structure in horizontal SSF wetlands, and near the bed surface in vertical SSFs and removed by similar mechanisms to suspended solid separation. This particulate matter will be converted to smaller particles by abiotic fragmentation, and then they can be hydrolysed by extra-cellular enzymes. These enzymes will be excreted by aerobic heterotrophs bacteria or fermentative bacteria. The smaller particles and the dissolved organic matter will be hydrolysed. Hydrolysis will generate lower molecular weight organic compounds that can be directly oxidized by the aerobic heterotrophs bacteria or fermentative bacteria (Garcia and Corzo, 2008). Figure 12: Simplified scheme of organic matter removal processes (Garcia and Corzo, 2008) Natural Systems for Wastewater Treatment in Warm Climate Regions Page 35 Moreover, it must be taken into account that some of the dissolved matter will be retained by adsorption onto the granular bed or the organic particles. In the horizontal systems, aerobic degradation can occur near the water surface and around the roots, but the oxygen released by the roots is not enough to remove completely the organic matter. Hence the predominant metabolic pathways are most likely anaerobic. The fermentative bacteria produce fatty acids and alcohols, which are substrate for the sulphur-reducing microbes and methanogens (both anaerobic). In the vertical systems, oxygen has been found along the bed depth, what suggests that aerobic degradation is the predominant metabolic pathway since the existence of oxygen in the bed inhibits any anaerobic reaction (Garcia and Corzo, 2008). Residual effluent from vertical systems is likely more consistent than that form the horizontal system, which shows that the aerobic conditions have better performances than anaerobic ones. Under both conditions, the removal performances vary between 75-90% producing outlet effluents with a BOD5 concentration below 20 mg/L (Garcia and Corzo, 2008). Moreover, residual effluent from SSF systems are likely more consistent than that from the FWS system because of the present of less plant matter in the water column (U.S. EPA, 2000). 7.4.3. Mechanisms for Nitrogen Nitrogen compounds are among the principal constituents of concern in WW because of their role in eutrophication. The wetland nitrogen cycle is very complex, and control of even the most basic chemical transformations of this element is a challenge in ecological engineering (Kadlec and Wallace, 2008). In WW, it is common to find nitrogen, normally in ammonia and organic nitrogen form. Nitrite and nitrate forms concentration are not usually significant (Garcia and Corzo, 2008). Organic nitrogen in WW includes proteins, peptides, nucleic acids and urea; these may be found in both soluble and particulate forms, while the other nitrogen species are water soluble. NH4-N may be found in the un-ionized form, NH3, or the ionized form NH4-, depending on the water temperature and pH. Due to normal wetlands conditions (pH=7 and 25oC), the ionized form is predominant (U.S. EPA, 2000). Figure 13: Common nitrogen species present in WW Natural Systems for Wastewater Treatment in Warm Climate Regions Page 36 The discharge of nitrogen to surface and groundwater sources is of concern for a number of reasons. Excessive accumulation of nitrogen will cause eutrophication, high concentration of ionized ammonia species are toxic to fish and other aquatic life, while nitrate and nitrite nitrogen are a public health concern (U.S. EPA, 2000). That is why is so important to evaluate the removal performances of all the species of nitrogen. Because of toxicity of un-ionized ammonia in receiving aquatic ecosystems, this nitrogen species is often singled out for regulation. The fraction of un-ionized ammonia depends upon water temperature as well as total dissolved ammonia (Kadlec and Wallace, 2008). Physical Separation Even though physical separation is not the main removal process, there are a number of separation processes that will affect nitrogen species in wetlands. Organic nitrogen associated to suspended solids may be removed by the same processes described earlier for the removal of TSS. Sorption of both particulate and soluble organic nitrogen and ammonia, because the positive charge, may occur on biofilms but this is a reversible process and as soon as the local conditions change, the organic nitrogen will be released again (Garcia and Corzo, 2008). As it was mentioned before, un-ionized ammonia (NH3) concentration is low at neutral pH, but during photosynthesis in open water zones pH may rise and its concentration too (U.S. EPA, 2000). If surface turbulence is high due wind action, un-ionized ammonia can be volatized. Biological Separation Almost half of the municipal WW nitrogen content is in the ammonia and organic nitrogen form. In wetlands, the main removal mechanism is a microbial process, which consists of a nitrification followed by a denitrification. However, nitrification can be also followed by plants uptake. In wetlands, the nitrogen cycle is coupled with the carbon cycle, mainly through the denitrification process. Below, these three processes are thorough described. NH4 NO2NO3NO N2O N2(g) - Nitrification Nitrification is the principal transformation mechanism that reduces the concentration of NH4-N in the water column. In the presence of dissolved oxygen (DO), microorganisms in the water column or onto the biofilm may convert ammonia to nitrite and nitrate nitrogen in a two-steps process. Ammonia oxidation is carried out by autotroph bacteria under aerobic conditions, with ammonia as the electron donor and oxygen as the electron acceptor. Nitrification process requires 4.6g of O2 per 1g of Natural Systems for Wastewater Treatment in Warm Climate Regions Page 37 ammonium nitrogen to oxide it to nitrate and 7.14g of alkalinity as CaCO3 is consumed, in order to maintain the proper pH. (1) 2NH4 + 302 2NO2- + 2H2O + 4H+ (2) 2NO2- + 02 2NO3Nitrate is not immobilized by soil minerals and remains in the water column or in the pore water of the sediments. It may be absorbed by plants or microbes in assimilatory nitrate reduction (converted to biomass via ammonium) or may be consumed by heterotroph bacteria and converted to nitrogen gas (denitrification). The presence of heavy metals in the water column may inhibit completely the reaction. The optimal pH range observed for nitrification is between 7.2 and 9 (Metcalf and Eddy, 1991). - Denitrification Denitrification or nitrate reduction is carried out by heterotroph bacteria under anoxic conditions, with organic carbon as the electron donor and nitrate as the terminal electron acceptor (U.S. EPA, 2000). The reaction occurs in the absence of oxygen and requires an organic carbon source; the minimum carbon to nitrate-nitrogen would be about 1g C/g NO3-N. The products of the reaction will be N2 and N2O gases which will exit the wetland. (3) CH3COOH + 2NO3 5/2 CO2 + N2 + 2OH + + 3/2H2O The process is temperature and pH dependent (U.S. EPA, 2000). A pH below 5 and an oxygen concentration above 0.3-1 mg/l in the water column may inhibit completely the reaction. - Plant Uptake (Assimilation) Wetlands plants can remove nitrogen by assimilating ammonia or nitrate as an important part of their metabolism, and convert it to biomass. They can reduce inorganic nitrogen forms to organic forms that are used for plant structure. During the growing season, there is a high rate of nitrogen uptake by emergent and submerged vegetation from the water and sediments. Nonetheless, during senescence, nitrogen is released to the water column, that is why is recommended lopping the vegetation before the senescence. Natural Systems for Wastewater Treatment in Warm Climate Regions Page 38 7.4.3.1. Nitrogen in Free Water Surface In FWS systems, each of the previously mentioned processes occurs but in different zones. Nitrification requires DO and therefore, there are limited areas of the wetlands where oxygen is available. There may be some nitrification occurring next to plants rhizomes were oxygen leaks from the plants, as well as, in the water column near the surface in open areas. The important variable in sizing an open water zone of FWS systems for nitrification is organic and nitrogenous loading because nitrification will not start until majority of the organic compounds have been removed. Thus, nitrification in the water column would not be expected in the initial settling zone. Under anaerobic conditions and in presence of organic matter, microbes associated with biofilms or suspended in the water column may convert nitrates to nitrogen gases via denitrification. Some nitrate will also diffuse into the sediments where it is available for plant uptake or can be denitrified as well. In open water zones of FWS systems, elevated pH and water temperature may enhance the NH3-N volatilization (U.S. EPA, 2000). Generally, designers have to be concerned with achieving removal by nitrification. This may be achieved at low loading (oxygen demand) with sequencing closed and open wetland areas (US EPA, 2000). Temperature affects both nitrification and denitrification, and performance rates can significantly decrease during the cooler months. 7.4.3.2. Nitrogen in Subsurface Flow Wetlands As described in section 7.4.2.2 Organic Matter in Subsurface flow wetlands, depending on the direction of the flow, the local conditions differ. In vertical wetlands, aerobic conditions prevail while in horizontal wetlands anaerobic conditions are more common. In vertical systems water flows pulsed. This operational regime provides alternating oxidizing and reducing conditions in the substrate, stimulating nitrificationdenitrification (Moshiri, 1993). However, in practice, it has been observed that in vertical wetlands aerobic conditions are more prevalent, that is why they achieve high performances in the conversion of ammonia to nitrate. In general, the nitrification is total. Denitrification permits remove the nitrate formed during the nitrification by converting it to nitrogen gas. However, this reaction only occurs under anoxic conditions and the presence of organic matter because this reaction is carried out by heterotroph bacteria. That is the reason why vertical wetlands are usually combined with horizontal wetlands. In horizontal wetlands the oxygen transfer is low and there Natural Systems for Wastewater Treatment in Warm Climate Regions Page 39 are few aerobic zones, so the nitrification process is not significant. But, they achieve high performance in denitrification as long as there is enough organic matter. It has been observed, that in horizontal wetlands, the processes of nitrification and denitrification occur coupled, thus, the formed nitrate is quickly reduced and converted to nitrogen gas (Garcia and Corzo, 2008). As described in the FWS systems, the ammonia can also be adsorbed, but this is a reversible process and as soon as, local conditions change, it will be release into water again. Plants can also remove the nitrogen by assimilating the ammonia or nitrates, and incorporate it to the biomass. But during the senescence, nitrogen can be release to the wetlands. Volatilization is not significant (Garcia and Corzo, 2008). 7.4.4. Mechanisms for Phosphorus Phosphorus occurs in natural waters and WW mainly as phosphates. They may be in solution or particulate form. Organic phosphates are produced mainly by biological processes, while inorganic phosphates come from fertilizers. The removal mechanisms can be biotic o abiotic. Biotic includes the assimilation by plants or microorganisms, while abiotic is mainly adsorption by the granular bed. Indeed, it is estimated that the proportion of phosphorus uptake by microflora and microfauna can be about 50% (Richardson, 1985). Particulate phosphate is usually associated to suspended matter, then, by removing suspended solids, particulate phosphate will be removed. The removal of phosphates is quite complex in CWs, including FWS and SSF systems. Experiences of the last three decades indicate that FWS wetlands can fulfil a useful role in phosphorus reductions in many situations. Improvements in water quality for secondary and tertiary effluents are possible, but there is perhaps an even greater role in controlling nutrients in urban and agricultural runoff. In contrast, SSF wetlands are rarely designed with phosphorus retention as a primary performance objective (Kadlec and Wallace, 2008). In general, performance rates are below 20% (Garcia and Corzo, 2008). 7.4.4.1. Phosphorus in Free Water Surface In FWS, soluble phosphates may be sorbed onto plant biofilms in the water column, or onto the wetland sediments. The exchange of soluble phosphates between sediment pore water and the overlying water column by diffusion and sorption is the major pathway (U.S. EPA, 2000). In the sediment pore water, organic form may be precipitated as the insoluble ferric, calcium and aluminium phosphates or adsorbed Natural Systems for Wastewater Treatment in Warm Climate Regions Page 40 onto clay particles. The precipitation as calcium phosphates occurs at pH values above 7 and may occur within the sediment pore water or in the water column near the phytoplankton. The sorption of phosphorus on clays involves negative charged phosphates and positive charges clay, and the substitution of phosphates for silicates in the clay matrix. However, phosphate can be released from the metal complex depending on the redox potential (Eh) of the sediment. Under anoxic conditions, phosphates may also be released from ferric and aluminium phosphates by hydrolysis. Plant uptake will only occur with dissolved inorganic phosphorus. That is why, dissolved organic phosphate and insoluble inorganics and organic phosphate may be transformed to a soluble inorganic form. This transformation may take place in the water column by suspended microbes and biofilms on the plants. Plant uptake is rapid, and following plant death, phosphorus may be quickly recycled to the water column or deposited in the sediments. Uptake by macrophytes occurs in the sediment pore water by the plant root system. The absence of vegetation lessens removal capability (Kadlec and Wallace, 2008). Temperature and vegetation growth patterns are two factors that may modify phosphorus uptake over the course of a year. Water temperature would be expected to modify microbial processes, which are involved in phosphorus uptake. However, the examination of wetland phosphorus removal data for warm climates shows only minimal seasonal effects. Figure 14: Phosphorus removal rate constants for the Orlando Easterly Wetlands, Florida (Kadlec and Wallace, 2008) 7.4.4.2. Phosphorus in Subsurface Flow Wetlands Removal of phosphorus occurs mainly as a consequence of adsorption and precipitation with aluminium, iron, calcium and clay minerals in the bed matrix. It has been noticed that phosphorus concentrations produced in SSF wetlands are a function of three primary variables: area, HLR and influent concentrations. However, Natural Systems for Wastewater Treatment in Warm Climate Regions Page 41 the sorption characteristics of the media are important factors that can be dominant mechanisms in treatment performance over the initial stage of operation until the sorption capacity is saturated (Kadlec and Wallace, 2008). At the start-up of the CWs, the phosphate removal will be high owing to the initial reaction with the soils of the wetland, but later on this rate is reduced quickly. This is because the clean granular bed has adsorption capacity but this is lost quickly. In vertical SSF, the wet and dry periods may enhance the fixation of phosphorus in the matrix (Cooper et al., 1996). Total phosphorus removal efficiency increases with higher input concentration and with higher HRTs (W.P.C.F., 1990). Currently, there is a lot of research coming on to develop new mechanisms to remove phosphates, but for the moment, it seems that the best mechanism is the precipitation of phosphates by adding aluminium sales. However, using iron sales to precipitate phosphates can lead to back colour water (Garcia and Corzo, 2008). 7.4.5. Mechanisms for Pathogens Pathogens are present in untreated domestic WW as well as in runoff waters from animal sources. Pathogens including helminths, protozoans, fungi, bacteria, and viruses are a great concern in assessing water quality. The density of these organisms in raw WW varies geographically. In order to evaluate the removal of pathogens, indicator organisms are used. The most common indicators of level of pathogen contamination are the faecal coliforms. Faecal streptococci analysis may also be used as an additional indicator of faecal pollution. Separation of pathogens and indicators, from the water column does not mean that the organisms are no longer viable. They may be released from the matrix to the water column and become available again. The true removal of pathogens is only achieved by making them nonviable. The efficiencies of conventional treatment technologies that reduce pathogens have been studies thoroughly and WWTPs regularly add processes to accomplish necessary removals (Metcalf and Eddy, 1991). The most common disinfection processes are chlorination, ozonation and ultraviolet irradiation. Meanwhile, natural treatment technologies have the potential to reduce populations of enteric pathogens because of natural die-off rates and hostile environmental conditions. Wetlands have been found to reduce pathogen populations with varying but significant degrees of effectiveness. Natural Systems for Wastewater Treatment in Warm Climate Regions Page 48 8. WASTE STABILIZATION PONDS 8.1. HISTORY AND INTRODUCTION Ponds have been used for centuries to store and treat animals and household WW (Gloyna, 1971). However, only within the last sixty years have specific design criteria been developed in terms of volumetric requirements, organic loadings rates, and HRT. Today, over 8,000 WSPs, comprising more than 50 percent of the WW treatment facilities in the United States are stabilizations ponds. In Europe, they are used to treat WW generated by small communities, while in New Zealand, Australia and Africa there are large pond systems (Mara, 2003). Over the last years, they have been without doubt the most important method of sewage treatment in warm climates when required land is available and the climate conditions are favourable (Mara, 1976). WSPs are large shallow basins in which WW is treated by entirely natural processes involving algae and bacteria (Mara, 1976). The main objective of WSPs is the reduction of the influent organic matter concentration (Von Sperling, 2007). Since these processes are not aided by man, the rate of oxidation is rather slow, thus, long HRT are required even under the favourable conditions in tropical regions (high temperatures and solar radiation intensity). The long HRT for the removal of organic matter has an important indirect advantage: WW remains for a sufficient long period to achieve complete removal of helminthic eggs and a high removal efficiency of faecal coliform, ensuring a high hygienic quality final effluent (Von Sperling, 2007). There are three major types of ponds: anaerobic ponds (AnP), facultative ponds (FP) and maturation ponds (MP). An anaerobic pond is essentially a digester, an aerobic pond is one in which aerobic bacteria break down the wastes and algae, through photosynthetic processes, provide sufficient oxygen to maintain an aerobic environment. Finally, the main function of the MP is to reduce the number of diseasecausing microorganism, and it may also be used to rear fish. In order to obtain a good performance and, at the same time, to minimize the HRT, various WSPs configurations are combined, and they operate in series. The first pond which receives that raw WW with high BOD5 load may become predominantly anaerobic, that is why they are called AnP, mainly removing the organic matter by settling on the pond bottom. The AnP is followed by a partially aerobic pond, named FP and finally in order to obtain a higher quality effluent there is the MP which is predominantly aerobic. Natural Systems for Wastewater Treatment in Warm Climate Regions Page 49 Notwithstanding, over the last decades, a large number of pre-treatment high rate anaerobic treatment systems have been implement in order to enhance the performance and quality of the outflow, specially UASBs (Campos, 1999). When anaerobic pre-treatment is applied, the concentrations of organic matter and TSS are drastically reduced and the removal of the residual concentrations in ponds becomes easier, shortening so much the HRT that the needed time to remove the pathogenic organisms and/or nutrients is sometimes bigger (Von Sperling, 2007). Thus, the ponds are referred to as secondary WSPs. The major disadvantage of ponds is that they require much larger areas of land than others forms of WW treatment. However, in many countries, especially tropical developing countries, this is rarely a disadvantage since sufficient land is normally available at relatively low cost. Some advantages of ponds are (Mara, 1976):  They can achieve any required degree of purification at a low cost and the minimum of maintenance by unskilled operators  As shown in table 1, WSPs and CWs are the cheapest form of sewage treatment, including construction and O&M costs.  The removal of pathogens is greater than that in other natural methods for WW treatment  The effluent from a series of three ponds usually contains less than 5,000 FC/100 ml, whereas the final effluent from a conventional treatment (humus tank effluent) typically contains about 5,000,000 FC/100 ml. Cysts and ova of intestinal parasites, which are commonly present in conventional effluents, are not found in MP effluents (Mara, 1976).  They can cope with organic and hydraulic shock loads  Long HRT ensure that there is always enough dilution available for short shock overloads.  They can effectively treat a wide variety of industrial and agricultural wastes  Wastes which are biodegradable have been successfully treated. For strong waster, AnPs are used, and in the event of heavily polluted sewage, anaerobic pre-treatment is utilized.  They can be easily re-designed so that the degree of treatment is readily altered  By designing the pond outlet structure, the top water level can be varied. Hence, the retention time too, and the degree of treatment is altered.  The method of construction is such that, if at some future date the land is required for some other purpose, it is easily reclaimed Natural Systems for Wastewater Treatment in Warm Climate Regions Page 50  All that is required is the removal of the inlet and outlet structure and level the ground.  The algae produced in the pond are a potential source of high-protein food, which can be conveniently exploited by fish farming.  Fish have been successfully grown in MPs. The sale of fish can bring in substantial revenue (Mara, 1976). 8.2. CLIMATE, PHYSICAL AND BIOLOGICAL FACTORS: - Climatic factors Physical, chemical and biochemical reactions that occur in WSPs are highly dependent on temperature (Rodríguez, 2008). The rate of degradation increases with temperature. The absorption of solar radiation plays a major part because it has an influence on: WW temperature, the photosynthetic activity and the removal of pathogens (CENTA, 2008). Light is essential to photosynthetic activity. As light intensity varies along the year, the algae growing rate varies too. This phenomenon has too effects: the variation of DO and pH in the water column (U.S. EPA, 2001). The wind has an important role because it leads the water column to be mixed, and ensures a uniform distribution of BOD, oxygen dissolved, bacteria and algae, so it enhances the degree of stabilization. The absence of mixing leads to stratification (CENTA, 2008). - Interaction between Bacteria and Algae In aerobic ponds, the presence of both algae and bacteria is essential for the proper functioning of the ponds (U.S. EPA, 2011). Bacteria break down the complex organic components into simple ones, which are then available for uptake by the algae (U.S. EPA, 2011). In turn; algae produce the needed oxygen for the survival of the aerobic bacteria. - Biochemistry of the ponds In WSPs, the carbonate buffering system has an important role. Its equilibrium is affected by the rate of algal photosynthesis. In photosynthetic metabolism, CO2 is removed from the dissolved phase, decreasing the hydrogen ion and increasing the pH. Because of the close correlation between pH and photosynthetic activity, there is a diurnal fluctuation in pH when respiration is the dominant metabolic activity (U.S. EPA, 2011). Natural Systems for Wastewater Treatment in Warm Climate Regions Page 51 The intensity and spectral composition of light penetrating a pond surface affects significantly the microbial activity. In general, activity increases with the increasing light intensity until it becomes light saturated (U.S. EPA, 2011). The quality and quantity of light penetrating the pond depend on the presence of dissolved and particulate matter. The organisms as algae, contribute themselves to water turbidity, limiting the depth of light penetration. Temperature at or near the surface determine the aquatic species. The major source of heat is solar radiation and there is a temperature gradient with depth (U.S. EPA, 2011). However, there is another heat source, which is the temperature of the influent water. In sewerage systems, the influent temperature is higher than that in the ponds. Thus, ponds may be prone to streaming, that is why is important a proper mixing (U.S. EPA, 2011). - Pond nutritional requirements In order to function as designed, the WW has to provide enough nutrients for the microorganism to grow and populate the system adequately. These nutrients include nitrogen, needed to algae uptake and bacterial action, phosphorus, which is most often the growth-limiting nutrient in aquatic environments, sulphur and carbon (CENTA, 2008). 8.3. DESIGNS OF STABILIZATION PONDS: Ponds are designed to enhance the growth of natural ecosystems, and they can be anaerobic, aerobic or FP, which is a combination of the two ones. The WSP system may comprise one pond only (FP) or several types of pond in series (AnP, FP and MP), even in parallel operation. There are many possible pond layouts. Figure 15: Typical pond layouts systems (adapted from Gloyna, 1971) Natural Systems for Wastewater Treatment in Warm Climate Regions Page 52 8.3.1. Anaerobic ponds (AnP) These ponds are designed to receive such a high organic loading and they are completely devoid of DO. They are neither aerated nor mixed. They are used to pretreat strong sewage which has high solids content. AnPs reduce the BOD load on the FP and change the nature of the settleable solids in the sludge layer, which will have a reduced fermentation potential. Anaerobic breakdown in septic tanks, UASB, and AnPs appears to be identical. The solids settle to the bottom where they are digested anaerobically, while the partially clarified supernatant liquor is discharged into the FP for a further treatment. During the anaerobic degradation process, there two main stages, each one carried out by a specific group of bacteria, the acid-forming and the methanogenic bacteria. The successful operation of AnP depends on the delicate balance between the acidforming bacteria and the methanogenic bacteria. Thus, a temperature higher than 15oC is needed and the pond pH must be over 6 (Mara, 1976). Ideally, temperature should be maintained within the range of 25-40oC, and the pH value should range from 6.6 to 7.6 (U.S. EPA, 2011). Under these conditions, sludge accumulation is minimal. Because AnPs are deep and generally have a relatively longer HRT, solids may settle, retained sludge is digested and organic matter concentration is reduced. Raw WW enters near the bottom of the pond and mixed with the active microbial mass in the sludge blanket. - Depth: AnPs are usually deeper than the other type of ponds; common depths are about 2-4 meters. Higher depths prevent from atmospheric oxygen diffusion. - Retention times: AnPs have detention times of 5-50 days. In the tropics, a liquid detention time of 1-5 days is recommended; longer detention may cause the upper layers of the pond to become aerobic (Gloyna, 1971). - Hydraulic surface loading: is a particularly important parameter affecting sedimentation. Hence, this rate has to be lower than the settling velocity of solids and pathogens or aggregates of pathogens. So, the hydraulic surface loading should be less than 2 m/d, which is approximately the settling rate of helminthic ova (Shilton, 2005). - Microbiology: anaerobic microorganisms convert organic materials into stable products, such as CO2 and CH4. The degradation process involves two separate but interrelated phases: acid formation, by “acid formers” bacteria, and methane production, by “methane formers” bacteria. During the acid phase, bacteria convert complex organic compounds to simple organic compounds, Natural Systems for Wastewater Treatment in Warm Climate Regions Page 53 mainly short-chain volatile organic acids. Then, bacteria convert the short-chain organic acids to acetate, hydrogen gas and CO2. Finally, those components are converted into methane by methanogenic bacteria. CH3COOH CH4 + CO2 CO2 + 4H2 CH4 + 2H2O When the system is working properly, these two phases of degradation occur simultaneously in dynamic equilibrium. However, the rate of degradation can be affected by the fluctuations of temperature and pH; even though the performance of acid-forming bacteria is the more tolerant to pH variation (U.S. EPA, 2011). - Loading: “acid formers” bacteria do not cope well with shock loads. - Mosquito breeding: in order to prevent the mosquito breeding, the pond must be kept free of vegetation. However, during the winter months when temperature decreases and removal too, it may appear an unsightly thick scum that promotes the fly-breeding. This problem can be overcome by increasing the maintenance (Gloyna, 1971). In AnPs, 80-90% of BOD5 removal can be expected, the sludge removal is rarely needed, and the energy requirements to run the plant are low or none. The main disadvantages of these ponds are the odour that they give off, and the extra maintenance that they require. The biochemical reactions in an AnP produce hydrogen sulphide and other odorous compounds. The relationship between the odour development and organic loading is now well understood, and can be minimised at the design stage. 8.3.2. Facultative ponds (FP) These are the most common ponds. They are usually used to treat the settle effluent from septic tanks and anaerobic pre-treatment ponds. The term “facultative” refers to a mixture of aerobic and anaerobic conditions; the aerobic conditions are maintained in the upper layers while anaerobic conditions exist towards the bottom. Figure 16: Typical cross-section of a FP (Gloyna, 1971) Natural Systems for Wastewater Treatment in Warm Climate Regions Page 54 Aerobic treatment processes in the upper layer avoid odour problems, and provide nutrients and some BOD removal. Meanwhile, anaerobic fermentation processes, such as sludge digestion, denitrification and some BOD removal take place in the lower layer. Most of the oxygen required to keep the aerobic conditions in the upper layer, is supplied by the photosynthetic activity of the algae which grow naturally in the pond, the other amount, comes from re-aeration through the surface. Indeed, the amount of algae is so high that the ponds are green in colour. The bacteria in the pond use the oxygen produced by the algae to oxidize the organic matter. The key to successful operation of this type of pond is the O2 production. One of the major end-products of bacterial metabolism is carbon dioxide which is used by the algae during photosynthesis since the demand for CO2 exceeds its supply from the atmosphere. Thus, there is an association of mutual benefit, symbiosis, between the algae and bacteria (Mara, 1976). Figure 17: Symbiosis of algae and bacteria in FP and MP (Mara, 1976) Since photosynthesis is a light-dependent activity, there is a diurnal variation in the amount of DO present in the pond and a similar fluctuation in the level of the oxypause, the point below the surface at which the DO concentration becomes zero, occurs. The pH also follows a daily cycle increasing with photosynthesis to a maximum which may be as high as 10. This happens because at peak demand algae remove CO2 from the solution more rapidly than it is replaced by bacteria respiration. These high pH conditions are favourable for ammonia removal via volatilization. The O2 in the upper layers is used by aerobic and facultative bacteria to stabilize organic matter. Anaerobic fermentation which takes place in the absence of oxygen is the dominant activity in the bottom layer of the pond. Natural Systems for Wastewater Treatment in Warm Climate Regions Page 55 - Depth: on the one hand, depths less than 1 meter may contribute to the emergence of vegetation and this must be avoided, otherwise, the pond becomes an ideal breeding ground for mosquitoes. On the other hand, with depths greater than 1.5 meters, the oxypause is too near the surface and the pond is predominantly anaerobic, which is undesirable (Mara, D., 1976). - Sludge layer: as the WW enters the pond, most of the solids settle to the bottom to form a sludge layer. At temperatures >15oC, intense anaerobic digestion of the sludge solids occurs; as a result, the thickness of the sludge is rarely more than about 250 mm and often much less. Desludging is required once every 10-15 years. At temperatures >22oC the evolution of methane gas is sufficiently rapid to buoy sludge particles up to the surface, forming a mat. This must be removed, together with any other floating debris, so they do not prevent the penetration of light into the photic zone, which usually comprises only the top 150-300 mm (Mara, 1976). - Climatic influences: a warm climate is ideal for pond operation. Solar radiation is intense and as a result, pond temperatures are high and there is enough intensity of light. The long daylight hours enable algal photosynthesis to occur for extended periods and so provide a reserve of DO for using during the night. However, there is usually a month of seasonal cloud cover and the light intensities are enough for algae activity, but not enough for algal and bacterial growth. That is the reason why the mean temperature of the coldest months is usually used as the design temperature (Mara, 1976). - Mixing: wind and heat are the two factors which influence the degree of mixing that occurs in a pond. Mixing minimizes the stagnant regions and it ensures reasonably the uniform vertical distribution of BOD, algae and oxygen and ensures that the non-motile algae are brought into the photic zone. In tropical areas when the wind velocity is low, the differential heating is the cause of mixing. In the absence of mixing thermal stratification quickly occurs. The warm upper layers are separated from the cold lower layers by a thin static region of abrupt temperature change. The non-motile algae settle and instead of producing oxygen, they exert an oxygen demand, creating quickly anaerobic conditions (Mara, 1976). - Retention time: recommended detentions vary from 5-50 days in warm climates. The main advantages include infrequent need for sludge removal, effective removal of settleable solids, BOD5, pathogens and faecal coliforms. They are easy to operate and require little energy. However, the main disadvantage is the higher sludge accumulation. Natural Systems for Wastewater Treatment in Warm Climate Regions Page 56 8.3.3. Maturation ponds (MP) MPs are used as a following stage to FPs; they maintain DO throughout their entire depth. Their main functions are the destruction of pathogens and provide a highquality effluent. They have shown to be one of the processes more efficient in the pathogens destruction (Yánez, 1995). The principal factor in the design of MP is detention time, but for efficient reduction of the pathogens it is essential that the pond is arranged in series (Gloyna, 1971). Faecal bacteria and viruses die off reasonably quickly owing to what to the inhospitable environment. The cyst and ova of intestinal parasites have a relative density, and as a result of the long retention times they settle to the bottom of the pond where they eventually die. The removal of BOD5 in MPs is low (Mara, 1976). - Depth: MPs are wholly aerobic and are able to maintain aerobic conditions at depths from 0.3 meters up to 3 meters. However, the depth of MPs and FPs are the same, around 1-1.5 meters. This is because of the destruction of viruses is better in shallow ponds than in deep ones because light penetrates better. But on the other side, they are not too shallow in order to prevent aquatic plant colonization. - Mixing: it is often provided, keeping algae at the surface to maintain the maximum rates of photosynthesis and O2 production and supplying added nutrients to the surface. - Retention times: detention time is typically 2-6 days. These ponds are appropriated for treatment in warm, sunny climates; mainly because they are used to destroy pathogens by UV radiation. However, the effluent will contain high TSS unless the algae are removed. The retention time, as well as the number of ponds, is determined primarily by the degree of bacterial purification required. The effectiveness of MPs in removing pathogens is conveniently assessed by the removal of faecal coliforms. With a proper design, rates of removal achieved may be greater than 99.99 per cent (Mara, 1976). In order to produce an effluent with a BOD5<25 mg/l, it has been found that two MPs in series, each with a retention time of 5-7 days are required (assuming that the FPs effluent is less than about 75 mg/l) (Mara, 1976). Natural Systems for Wastewater Treatment in Warm Climate Regions Page 57 8.4. REMOVAL MECHANISMS IN STABILIZATION PONDS: 8.4.1. Mechanisms for Suspended Solids TSS present in the column water can be organic or inorganic mater. A fraction of these particles may settle out by their own weight, while other may settle because microorganisms may adhere to the surface of particles forming flocs that will settle out. This will lead progressively to an accumulation of sludge on the bottom of the ponds, susceptible to be degraded by microorganism than live within it. However, in aerobic ponds due to the symbiosis of algae and bacteria, some of the organic matter present in the water may be assimilated by the algae, increasing the content of suspended solids. In FPs, the removal rate of TSS varies over the year, being really low or even negative during the season of mass growth of algae (spring and summer). In MPs, the presence of protozoa and small crustaceous contributes slightly to the removal of particulate organic matter (CENTA, 2008). The occasional high concentration of TSS in the final effluent can be the major operational challenge for pond systems. The solids are composed primarily by algae and other pond detritus, not WW solids. These high concentrations usually occur during summer. In order to remove this TSS, different methods have been used, such as intermittent sand filters, recirculating sand filters, rock filters, coagulationflocculation and dissolved air flotation. Because of this dissertation is focused on natural treatment systems which require low or even no energy. It would be recommendable to use the intermittent sand filters which have demonstrate their capability of polishing pond effluents at a relatively low cost (U.S. EPA, 2011). Intermittent sand filters are similar to the practice of slow sand filtration in potable water treatment, as the effluent passes through the bed, TSS and other organic matter are removed through a combination of physical straining and biological degradation processes. The accumulation of matter finally clogs the surface of the filter and prevents effective infiltration. At that time, the bed is taken out of service and cleaned. The typical HLRs range from 0.37-0.56m3/m2/d, but could be lower is the TSS concentration exceeds 50mg/l. Algae removal is almost totally a function of the sand size used. Algal TSS may be used as a nutrient for use in agriculture or as a feed supplement (Grölund, 2002). Natural Systems for Wastewater Treatment in Warm Climate Regions Page 64 oxidising species that, in turn, damage internal targets within the cell or viral particle. This mechanism depends on the DO in the external medium. Mechanism 3 involves absorption of a wide range of ultra-violet and visible wavelength (400-700 nm) in sunlight by humic organic solids, causing direct damage the cell membrane. The activated photosensitizers react with oxygen to form highly reactive photo-oxidising species that, in turn, damage external targets, including the membrane of bacterial cells. This mechanism is also dependent on DO. Davies-Colley et al. (2000; 1999) showed that enterococci (part of the faecal streptococci group) were more rapidly inactivated by sunlight (by photo-oxidative mechanism) than E. coli, expect at elevated pH (above about 9) under which conditions accelerated inactivation of E. coli make it less persistent. Generally it appears that E. coli are the better indicator except at very elevated pH (>9.5) when this bacteria is more rapidly removed than enterococci and some pathogens (Shilton, 2005). Removal of infectious worm parasite eggs from domestic WW is especially important in developing countries where community infection levels are often high (Mara, 2001). Multiple-pond systems are capable of efficient removal of helminthic eggs mainly by the process of sedimentation to the sludge. Then, the sedimentation of these eggs transfers the concern from the water to the sludge. However, 100% removal efficiency is not always guaranteed. Protozoan pathogens are persistent in the environment owing to their formation of resistant cysts. Experimental data suggests however, that, despite their environmental resistance, protozoan cysts are effectively removed within WSPs (Shilton, 2005; CENTA, 2008). Although disinfection by WSPs is generally really good and much better than in mechanical treatment plants (George et al., 2002), the final effluent quality is still variable. Hence, some further disinfection treatment may be needed to meet the stringent standards. 8.4.6. Mechanisms for Heavy Metals: Heavy metals may be removed from WSPs by a variety of processes, including: sedimentation of solids, adsorption to algal/bacteria biomass and bottom sludge, bioaccumulation into algal/bacteria biomass, chelation and precipitation. Most heavy metals are associated with particulate matter and therefore, they settle out. Adsorption of heavy metals onto the surface of algae and bacteria cell is a rapid process. Adsorption involves attraction of the positively charged metal ions to the Natural Systems for Wastewater Treatment in Warm Climate Regions Page 65 numerous negatively charged sites on the surface of algae and bacteria cells, what results into the displacement of divalent or monovalent cations. Algae and bacteria are also known for their capacity to accumulate heavy metals since they are requires as essential micronutrients (Shilton, 2005). The accumulated metal ions are usually compartmentalized within the cell or converted to less toxic forms by binding or precipitation (Gadd, 1990). However, at high concentrations, they can inhibit the growth of algae and bacteria and may even cause death (Gadd, 1990). Moreover, many algae and bacteria release extracellular secretions that act as chelating agents. These chelating agents from complexes with free heavy metals ions and, hence bioaccumulation will be reduced as well as toxicity. However, these heavy metal chelates are only stable at high pH (Gale and Wixon, 1979). Heavy metals are most toxic in their free ionic form; therefore, toxicity decreases as pH is high due to the formation of insoluble precipitates (Rai et al., 1981). They may precipitate under both anaerobic and aerobic conditions. Under anaerobic conditions, heavy metals precipitate with sulphites. Whereas, under aerobic conditions and at high pH, heavy metal cations combine with anions such as hydroxide and phosphate (Rai et. Al, 1981; Shilton, A., 2005). There is little information on heavy metals removal in WSPs. Most removal occurs in primary ponds, AnP or FP, and is due to sedimentation of solids to which heavy metals are sorbed (Toumi et al., 2000; Shilton, 2005). The following table 7 summarizes the removal rates of the contaminants depending on the type of WSP. TSS BOD5 COD N P AnP 50-65 40-50 40-50 5-10 0-5 FP 0-70 60-80 55-75 30-60 0-30 MP 40-80 75-85 70-80 35-80 10-60 Table 7: Removal rates of contaminants (%) at the different types of WSPs (CENTA, 2008) 8.5. ALGAE CONTROL: Algal overgrowth is a matter for concern because it can cause the depletion of oxygen during the respiration phase, and may increase the TSS concentration in the pond final effluent. Algal overgrowth is prevalent in the areas where there are a high number of sunny days during the year, long HRT, shallow pond depths, abundant nutrients, warm water Natural Systems for Wastewater Treatment in Warm Climate Regions Page 66 and sunshine. The problem is that during the night, algae and aerobic bacteria will utilize oxygen during the respiration process, potentially depleting the DO in the column water and causing incomplete treatment. However, on the other side, the high concentration of algae at the surface will reduce sunlight penetration and may slow the growth rate. On the other side, control of algae in WW treatment ponds effluents has been a major concern throughout the history of the use of these systems. As it was mentioned in section 8.4.1 Mechanisms for Suspended Solids, algae grow in MPs increases the TSS in the final effluent. It has been established that few, if any, of the solids in the final pond effluent are faecal matter or material entering the pond system (U.S. EPA, 2011). This has led to a discussion about the need to removal algae from the effluent, because when algae die, settle out and decay, they do create some O2 demand on the receiving stream. Algae require light to grow, and as light penetration is reduced with increasing depth, so, increasing the depth of the MPs, up to 3-4 meters, the algae growth will be reduced. Without mechanical mixing, thermal stratification occurs in ponds, providing an excellent environment for algae to growth. Disturbing stratification, and reducing light transmission, will help to reduce the rate of growth (U. S. EPA, 2011). 8.6. ODOUR RELEASE AND CONTROL: The release of offensive odours from AnPs occurs when the volumetric loading on the pond is greater than 400 g BOD5/m3·d (Mara, 1976). Thus, even for a strong sewage (BOD5=1 000mg/l), odour release is unlikely to be a problem when the retention time is 5 days. However, a high concentration of sulphates in the water, especially if it is agricultural or industrial waste, may cause odour problems. In this case, odour control is required, and this may be achieved by: - Raising the pH of the pond to about 8, so most of the sulphide will exist as the odourless bisulphide ion, HCl-. - Recirculating the effluent from the FP or MPs to the AnP inlet in the ratio 1 to 6 (1 volume of the effluent, with higher DO, to 6 volumes of raw sewage) (Mara, D., 1976). This provides a thin aerobic layer at the surface of the AnP, which prevents odours from escaping into the air. - A cover may also use to contain odours. Natural Systems for Wastewater Treatment in Warm Climate Regions Page 67 8.7. UASBs VS. ANAEROBIC PONDS: Even though this dissertation is focused on a pond layout with an UASB as a primary treatment followed by a FP and MP, it is a matter of interest discuss if an AnP would be a better option. A 6h UASB achieves a 70% removal of BOD, but this is also achieved by a 1d AnP at 25oC (Mara, 2003). The UASB is clearly smaller: it has only one quarter of the volume of the AnP. However, it costs more to construct a 6h UASB in reinforce concrete, even reinforced brickwork, than it costs to construct a 1d AnP (Mara, 2003). Moreover, the saving in land area is insignificant when compared with the area of the secondary FP needed to treat the anaerobic effluent and the area of the drying beds for the UASB sludge. However, the AnPs may present greater odour problems. Hence, the choice of one or another primary treatment is not trivial. It may depend on the land available and surroundings, the design criteria, the budget and the experience of the client and duty holders. Natural Systems for Wastewater Treatment in Warm Climate Regions Page 68 9. WATER REUSE: A growing world population, the unrelenting urbanization, the increasing scarcity of good quality water resources and the rising fertilizer prices are the driving forces behind the accelerating upward trend in the use of WW, excreta and greywater for agriculture and aquaculture (WHO, 2008). The principal forces driving this increased use are:  Increasing water scarcity and stress  Expanding population with increasing environmental pollution  Recognition of the resource value of WW, excreta and greywater It is estimated that within the next 50 years, more than 40% of the world’s population will live in countries facing water stress or water scarcity (Hinrichsen et al., 1998). Indeed, in many cases, it is better to use WW, excreta and greywater in agriculture than to use higher-quality fresh water, because crops benefit from the nutrients they contain (WHO, 2008). Most population growth is expected to occur in urban and periurban areas in developing countries (United Nations Population Division, 2002). The reuse of WW will be an important component of a package of coping strategies in areas affected by such change. For agriculture use which includes irrigation of crops, sports fields and public parks, it has been established by the recommended guidelines for unrestricted WW use in agriculture of WHO, that the faecal coliforms concentration must be lower than 1,000 CFU/100 ml and the concentration of helminth eggs must be lower than 1 egg/litre. Meanwhile, for aquaculture use, the faecal coliform concentration has to be lower than 1 000 CFU/100 ml and none viable eggs (WHO, 2002). Natural Systems for Wastewater Treatment in Warm Climate Regions Page 69 10. CASE STUDY: Influence of heavy rain episodes on removal efficiency in three stage hybrid treatment wetlands The performance of an experimental hybrid CW pilot system was assessed during three months. Moreover, a heavy rain episode, a characteristic phenomenon of tropical climate regions, was simulated. The aim was to assess the appropriateness of this system for warm climate regions. The following sections show the description of the pilot system, as well as the results of the experiments carried out from June 2013 to September 2013. 10.1. DESCRIPTION OF THE PILOTE-SCALE TREATMENT WORKS: The experimental hybrid CW pilot system belongs to the Group of Environmental Engineering and Microbiology (GEMMA). It is located at the Department of Hydraulic, Maritime and Environmental Engineering (DEHMA) of the Universitat Politècnica de Catalunya, Spain. The pilot plant consists of: - Preliminary treatment: raw WW tank, fine screening and stirred tank. - Primary treatment: HUSB - Secondary treatment: two vertical SSF CWs, and one horizontal SSF CW - Tertiary treatment: one FWS CW All these different elements are set up within two skids of 11 m2 each one. Figure 19: Top view of the pilot-scale treatment works (adapted from Donoso, 2013) HUSB Vertical SSFs Stirred Tank Horizontal SSF FWS Natural Systems for Wastewater Treatment in Warm Climate Regions Page 70 Figure 20: Studied pilot-scale treatment plant (Donoso, 2013) 10.2. DESCRIPTION OF THE TREATMENT PROCESS: Urban WW was pumped directly from the nearest municipal sewer into the system by two pumps. Once in the system, WW was undergone a fine screening and poured into a 1.2 m3 polyethylene stirred tank, in order to prevent any sediment settling out. After that, WW was pumped into the HUSB by means of a peristaltic pump with an input flow rate of 800 l/d. Three flowmeters were installed at the entrance of the HUSB, the horizontal SSF and the FWS CWs. The HUSB had a nominal HRT of 5 hours for a design flow of 1200 l/d. Once in the HUSB, the organic load of the effluent was reduced in order to enhance the performance of the CW system. The HUSB is equipped with 9 taps positioned vertically in series, starting at a height of 48 cm from the bottom and located at a distance of 20 cm from the previous one. By this distribution, it is easy to regulate the level of the mud inside the reactor. To reduce the lag phase of the microorganism in the sludge and to accelerate the stabilization of the sludge layer, the HUSB was inoculated with secondary sludge from the wastewater treatment plant of Gavà (Catalonia, Spain). Fifty litres of sludge were inoculated two times. After that, the effluent flowed into a tank of 0.25 m3 that regulated the amount of water pumped to the vertical SSF wetlands by means of two pumps. The two vertical CWs operated alternatively in cycles of 3.5 days; this pulsed pumping was done in order to ensure aerobic conditions within the wetland. Moreover, each one has a metal tramex plate above the floor level and a number of holes to allow passive aeration of the bed. Each pressure pump fed each of the vertical CW. Both vertical CWs are identical, with a surface area of 1.5 m2. They have a feeding pipe, 0.10 meters above the surface of the bed, with 5 holes with diffusers that ensure a Natural Systems for Wastewater Treatment in Warm Climate Regions Page 71 360o radial horizontal water pattern. The water passed through the granular bed and was recollected at the bottom where flowed into a small tanks of 0.25 m3. This tank is covered to avoid light exposure and is necessary for the sampling of the effluent after this treatment stage. Figure 21: Cross section of the vertical SSF CW From this tank, WW was pumped by means of a peristaltic pump to the horizontal SSF wetlands; which has a surface area of 2 m2. Figure 22: Cross section of the horizontal SSF CW Finally, the horizontal SSF effluent is send to another 0.25m3 tank that allows sampling and then pumped to the tertiary treatment which is a FWS wetland with a surface area of 2 m3, achieving a high quality effluent ready to reuse. Natural Systems for Wastewater Treatment in Warm Climate Regions Page 72 Figure 23: Cross section of the FWS CW All CWs were planted with Phragmites australis. In the case of the FWS wetland, only third of the surface area was covered, in order to allow sunlight penetration and obtain mixed conditions. Figure 24: Hybrid CW System (Imhoff tank was remplaced by a HUSB reactor) (Avila et al., 2013) 10.3. HYDROLOGICAL PARAMETERS: 10.3.1. Under design conditions: The plant had been operating with clean water for almost one year, but it was not until March 2013 that the whole treatment plant started to work correctly. Since then sampling campaign and analysis started. The hybrid CWs system was evaluated under design conditions during nearly four months, the hottest months of the year, from June 2013 to September 2013. - Influent flow: 33 l/hour - Nominal HRT of the complete system: 24h Natural Systems for Wastewater Treatment in Warm Climate Regions Page 73 - Average OLR for vertical SSF: 54.7 g BOD5 m-2 d-1 (calculated considering the surface of vertical flow wetlands (3 m2)) - Average HLR for vertical SSF: 0.27 m/d - Average temperature: 23.4oC 10.3.2. Under heavy rain conditions: Monsoons are climatological phenomena associated to the weather patterns of tropical and sub-tropical continents. Summer monsoons are large-scale sea breezes which occur when the temperature on land is significantly warmer than the temperature of the ocean, which causes a heavy rain over the land. On September 2013, a heavy rainfall period was simulated. The WW was mixed with potable water, increasing the flow rate 10 times more than the normal influent. The treatment plant had to be adapted accordingly, and the two peristaltic pumps that feed the horizontal SSF and FWS were changed by two centrifugal pumps in order to meet the new input flow. During the Monsoon simulation, the pilot plant worked under and HLR of 330 l/h (33 litres of WW + 300 litres of potable water) during 1 h. The duration of the experiment was 10 h. The first sampling was made just before the beginning of the storm and immediately after, and then samples were taken every 1 hour during 9 hours. 10.4. SAMPLING STRATEGY: As mentioned before, the plant operated under an input flow of 33 l/hour and was monitored from June 2013 to September 2013. Previous assay were carried out from February 2013 to June 2013 and were compared with this last campaign in order to see any improvement in NH4-N removal rate with higher temperatures. Grab samples were taken one a week for the analysis for the following parameters: pH, DO, Eh COD, BOD5, TSS, NH4-N. Sampling points are shown in Figure 25. The sludge blanket within the HUSB reactor was sampled twice a week to ensure that VS concentration was lower than 10 g/l. Hence, the VS concentration was measured at each tap of the HUSB. Natural Systems for Wastewater Treatment in Warm Climate Regions Page 80 10.5.2. Under heavy rain conditions: This section shows the results of physical and chemical analysis obtained during the heavy rain episode test, when the plant operated under an input flow of 330 l/hour. The first sampling point was taken just before the start of the heavy rain episode simulation, so, under normal conditions. Between sample one (1h) and two (2h) the flow was increased ten times, simulating the heavy rain storm. Sample two (2h) was taken just after the storm episode simulation. 10.5.2.1. Chemical Oxygen Demand Figure 32 shows the evolution of COD concentration in each stage during the heavy rainfall campaign. As expected, in the stirred tank the COD decreased drastically because of the dilution of the raw WW with the income rainfall during the duration of the episode (from hour 1 to hour 2) and then, after the end of the episode (hour 2), it increased up to the normal conditions values. In the HUSB, the process was similar; it showed a reduction of COD concentration during the first 2 hours, and remained low until hour 6 when COD concentration rose again. This turning point happened 5 hours after the rainfall episode started, which matched up with the HRT of the reactor. The vertical SSF CWs showed a minimum in COD concentration, 50.97 mg/l, at hour 6, matching with the time the WW from the episode flowed from the CWs. The same happens in the horizontal SSF CW and FWS CW with a minimum concentration value of 7.48 mg/l and 11.44 mg/l, respectively. The SSF CW and FWS CW showed a relatively stable concentration during the whole campaign, which indicates the robustness of the system and their capability to cope with heavy rain episodes. Natural Systems for Wastewater Treatment in Warm Climate Regions Page 81 Figure 32: Evolution of the COD concentration in each stage of the system during the heavy rain campaign 10.5.2.2. Total Suspended Solids The evolution of TSS concentration in each stage of the system during the experiment is showed in Figure 33. The evolution is really similar to COD concentration. As expected during the rainfall episode (the interval 1-2), the TSS concentration decreased drastically in the stirred tank and HUSB. Then, the concentration in the stirred tank waved. Whereas, in the HUSB the concentration remained really low, below 25.75 mg/l, during 4 hours, increasing again at hour 6, matching up again with the HRT of the reactor. Vertical SSF CWs showed minimum values of TSS concentration after 6 hours, matching with the time the WW from the episode flowed from the CWs. Horizontal SSF CW and FWS CW showed again constant TSS concentration, indeed, really low values, below 6.33 mg/l and 1.92 mg/l respectively. 0 100 200 300 400 500 600 700 800 1 2 3 4 5 6 7 8 9 10 Concentration (mg/l) Evolution of COD concentration in each stage of the system Stirred tank HUSB Vertical SSF Horizontal SSF FWS Natural Systems for Wastewater Treatment in Warm Climate Regions Page 82 Figure 33: Evolution of the TSS concentration in each stage of the system during the heavy rain campaign 10.5.2.3. Ammonia-Nitrogen A drastic drop of NH4-N concentration occurred in the stirred tank and HUSB just after the rainfall episode started. Due to the dilution of the WW with the simulated rain it decreases from 22.75 mg/l to 2.45 mg/l in the stirred tank, and from 32.95 mg/l to 8.66 mg/l in the HUSB. After the end of the storm (2h), the NH4-N concentration in the tank returned to normal values. However, in the HUSB, values remained low during 4 hours after the start of the experiment, below 11.77 mg/l. Both vertical and horizontal CWs showed a peak concentration about the same hour 2, 17.61 mg/l and 7.12 mg/l respectively. Then, NH4-N concentration also rose in vertical CW at hour 4, 17.03 mg/l, but after the concentration decreased gradually. Whereas in horizontal CW concentration decreased. FWS CW showed constant concentrations during the whole campaign, below 6.66 mg/l. 0,00 20,00 40,00 60,00 80,00 100,00 120,00 140,00 160,00 180,00 200,00 1 2 3 4 5 6 7 8 9 10 Concentration (mg/l) Evolution of TSS concentration in each stage of the system Stirred Tank HUSB Vertical SSF Horizontal SSF FWS Natural Systems for Wastewater Treatment in Warm Climate Regions Page 83 Figure 34: Evolution of the NH4-N concentration in each stage of the system during the heavy rain campaign 10.6. DISCUSSION: 10.6.1. Discussion of the results under normal conditions: An overview of the general performance of the plant will be discussed in this section. First of all, it is important to point out that the sewer system from which the WW was collected, was located at high income residential area with a large number of schools. Thus, there may be a great variation on WW composition especially in summer, with a great amount of WW coming from gardens and swimming pools. In short, the composition of WW could not be defined and controlled; the water quality parameters of the influent had a high variability. In Annex I, the average values and standard deviations of the measured parameters in each stage of the treatment plant are shown. Average influent concentrations of COD and BOD5 were 226.24 ± 100.11 mg/l and 154 ± 45.06 mg/l respectively. These values are really low and far from the typical sewage concentrations, this could be caused by the fact that during the hottest months a great amount of WW came from garden’s irrigation and swimming pools and schools from the surrounding were closed. Another matter of concern is the fact that the removal rates for COD and BOD5 of the HUSB were -17.4% and -23.4%, so, the pre-treatment did not operate properly at all. The reason why to use a pre-treatment is to reduce the concentration of TSS, COD and BOD5, to improve the efficiency of the secondary and tertiary treatment and prevent from clogging, but instead of that, there was an increased in concentrations. Even so, CWs showed good removal efficiency. Vertical SSF CW showed a removal efficiency of 34.8% and 81.4% for COD and BOD5 0 5 10 15 20 25 30 35 1 2 3 4 5 6 7 8 9 10 Concentration (mg/l) Evolution of NH4-N in each stage of the system Stirred tank HUSB Vertical SSF Horizontal SSF FWS Natural Systems for Wastewater Treatment in Warm Climate Regions Page 84 respectively. Horizontal SSF CW provided about 64.5% and 64.7% removal of COD and BOD5 respectively. TSS mass removal of the HUSB was even worse, -85.34%, which is a clear warning that the HUSB is not working correctly, this mass was coming from the added sludge of the HUSB. According to Metcalf and Eddy, the normal value of TSS mass removal in physical pre-treatments may be about 50-60%. Vertical SSF CW showed a removal efficiency of 69.7% for TSS while horizontal SSF CW provided about 72.6% of TSS removal. Even so, the removal rate in FWS CW was 78.1% and the overall system reached about 96.6% of TSS removal. NH4-N concentration, as well as the above mentioned parameters, increased in the HUSB about 38.7%. However, the CW system showed a good removal efficiency especially the vertical SSF CW, 67.4%, and overall, the removal rate was about 90.8%. As noted before, this result was compared with the NH4-N removal rate of a previous campaign carried out at the same pilot-scale plant during the cool season by Amigó (2013). As expected, the removal rate during the warm season (90.8 %) was far higher than the one obtained during the cold season (78%). The high concentration of DO in the influent may be caused by the continuous aeration in the stirred tank before flowing to the HUSB. Once the WW was within the HUSB, DO concentration decreased because the consumption of oxygen to degrade organic matter, 2.13±0.92 mg/l, and then, as WW percolated through the bed of the vertical SSF CW, DO concentration increased, 2.95±1.89 mg/l, because the alternating aerobic and anaerobic conditions. The DO concentration in the final effluent is really low. It should be aerated before going to the receiving water. Fish needs DO≥5 mg/l. According to studies carried out by Vyzamal et al. (2008) and Barros et al. (2008), anaerobic digester provided a COD removal about 35-65%, about 35-65% removal of BOD5 and 50-90% removal of TSS. However, the results of this study are the complete opposite and indicate that the implementation of a HUSB reactor as a primary treatment did not enhance the treatment capacity of the system. Despite the malfunctioning of the anaerobic pre-treatment, overall treatment efficiency range from 90.8 to 96.6% removal for NH4-N, BOD5 and TSS, and 77.6% removal for COD. 10.6.2. Discussion of the results under heavy rain conditions: First of all, it has to be pointed out that the first flush and increasing OLR was not simulated due to technical limitations. Hence, the expected concentration curve of a real storm case did not occur. This curve is characterized by a peak of the concentrations occurring a little bit after the beginning of the storm, followed by a decrease of the concentrations (Avila et al., 2013a). In this case, all the water quality Natural Systems for Wastewater Treatment in Warm Climate Regions Page 85 parameters concentration suffered a drastic drop because of the dilution with potable water. In Annex II, the concentration values of the measured parameters in each stage of the treatment during the heavy rain campaign are shown. As it can be seen in Annex II, the influent COD concentration was about 300-400 mg/l. This value is quite higher than the one from normal conditions because by the time the heavy rain episode was simulated (19th of September 2013), schools were open again and people had come back from holidays or second homes. At the point 5 and 6, the vertical SSF CW showed an increase on COD concentration. This could be explained by the fact that the diluted WW was still in the HUSB, which had a nominal HRT of 5h, and the water collected in the vertical wetland was previous to the heavy rain episode. After this “breaking point” the performance of the vertical was the expected one. The average removal rates of the treatment system for COD, TSS and NH4-N were 83%, 99% and 80% respectively. It has to be mentioned that these values are not exact but an approximation, because to calculate the overall rate at each point it was not considered that in some units of the system the water was diluted and in others it was not, due to the different retention times of the stages. Indeed, a tracer experiment should be carried out to obtain the real HRT of each unit and for a better understanding of the results. It was also important to study the response of the HUSB to the heavy rain episode, especially the response of the sludge. Despite of the increased flow, the HUSB did not lose much sludge. It was measured the solids concentration of the effluent the day before and the day after the test, resulting 8.88 g/l and 7.12 g/l respectively, and the following week the concentration was again stable. To sum up, the removal rates of the treatment plant did not vary significantly from the ones obtained under normal conditions and the anaerobic digester-CW system showed a good efficiency during the experiment. The contaminants concentration seemed to return to the normal average around 7 hours after the rain episode for some units (i.e: HUSB, VF CW). On the other hand, for HF CW and FWS CW fairly constant concentrations were observed. It could be due to the higher HRT of the pilot plant compared to the duration of the experiment. As mentioned above, a tracer experiment should be carried out to obtain the real HRT of each unit and for a better understanding of the results. Finally, the system is robust and it can handle on heavy rain episodes, which makes it a suitable water treatment engineering solution for warm climate countries. Natural Systems for Wastewater Treatment in Warm Climate Regions Page 86 11. CONCLUSIONS AND RECOMMENDATIONS: The work carried out during the dissertation led to the following conclusions: - About the literature review of natural systems: Against a backdrop of restrictive directives in which small and rural communities have to treat their WW prior to discharge, this dissertation has tried to review and find the most suitable engineering solutions considering the environmental, aesthetic and cost aspects. The natural systems for WW treatment presented in this dissertation are feasible solutions to treat sewage from small communities and allows to its further reuse. Apart from a good removal rates, CW systems provide an additional value to the treatment because they allow recovering lost natural zones and ecosystems. However, FWS CWs could be prone to mosquito development. WSPs allow to removal organic matter, nutrients and pathogens from WW with minimal O&M costs. However, the main disadvantage is the substantial increase in algae content in the final effluent, which may need a further treatment such as sand or rock filters. Both natural systems show good removal rates for TSS, organic matter, and NH4-N. However, phosphorus mass removal is low, and the effluent could cause eutrophication in the receiving water. Phosphorus removal could be improved by adding salts or flocculants, however this increase considerably the costs. There may be a need to develop a low-cost technology to remove phosphorus content. The combination of anaerobic digesters and CWs or WSPs for the secondary and tertiary treatment of domestic WW is a recent and promising solution in developing countries. - About the experiment to assess the efficiency of a three stage hybrid treatment wetlands: The performance of pilot-scale treatment plant, which consisted of an anaerobic digester followed by a hybrid CWs system, during the warmest months of the year and its robustness under an extreme rainfall event was tested. Under an input flow of 800 l/d, the average values of the total mass removal rates were above 77.6% for all the contaminants even though the HUSB did not work properly during the period considered. During the heavy rainfall campaign, the total mass removal rates were even higher (above 80%). The system seemed to return to its normal average values 7 hours after Natural Systems for Wastewater Treatment in Warm Climate Regions Page 87 the rainfall episode for some units (i.e: HUSB, VF CW). For HF CW and FWS CW fairly constant concentrations were observed. Moreover, the sludge within the HUSB could handle on the increased flow. In short, the system showed a very good buffer capacity under extreme rainfall events. It was proved that the system can cope with the sharp fluctuations in flow to be treated. It can be concluded that the technology of CWs is a valid solution for WW treatment generated in small agglomerations of warm climate areas. Indeed, the experimental hybrid system showed to be highly efficient. The study has also allowed proposing the following recommendations:  During the design stage of FWS CWs, the potential hazard of mosquito development should be considered and minimized. That is why SSF CWs are preferred.  In WSPs, AnPs may also be a feasible option for primary treatment, instead of an UASB. Each project is unique, that is why the choice of one or another primary treatment is not trivial and should be evaluated.  In relation to the pilot-scale plant, it may be advisable to study the performance of the hybrid CWs system without the primary treatment. Moreover a tracer experiment should be carried out to obtain the real HRT of each unit and for a better understanding of the results obtained from the Monsoon simulation. Natural Systems for Wastewater Treatment in Warm Climate Regions Page 88 REFERENCES: 1. Abeliovich, A. and Azov, Y., 1976. Toxicity of ammonia to algae in sewage oxidation ponds. Appl. Environ. Microbiol. 31, 801-806 2. Álvarez, J.A., et al., 2008. Anaerobic digesters as a pre-treatment for constructed wetlands. Ecological Engineering 33(1), 45-67 3. Alvord, H.H and Kadlec R.H., 1995. The interaction of atrazine with wetland sorbents. Ecological Engineering 5(4) 4. Alvord, H.H and Kadlec R.H., 1996. Atrazine fate and transport in the Des Plaines Wetlands. Ecological Modelling 90(1) 5. Amigó, N., 2013. Influence of hydraulic loading rate on treatment efficiency and water quality for reuse in a three stage hybrid constructed. UPC 6. Avila, C. et al., 2013. Three stage hybrid constructed wetlans system for wastewater treatment and reuse in warm climate regions. Bioresource Technology 7. Ávila, C., et al., 2013a. Integrated treatment of combined sewer wastewater and stormwater in a hybrid constructed wetland system in southern Spain and its further reuse. Ecological Engineering 50, 13-20. 8. Barros, P., et al., 2008. Performance of an anaerobic digester-constructed wetlands for a small community. Ecological Engineering, 33, 142-149 9. CENTA, 2008. Manual de depuración de aguas residuales urbanas. 10. Chevalier, P. and De la Noüe, J., 1985. Wastewater nutrient removal with microalgae immobilized in carrageenan. Enz. Microb. Technol. 7, 621-624 11. Craggs, R.J. et al., 2000. Nitrification potential of attached biofilms in dairy farm waste stabilization ponds. Water Science and Technology, 42(10-11), 195-202 12. Crites, R. W. et al., 2006. Natural WASTEWATER treatment systems. CRC Press, Boca Raton 13. Cooper, P.F. et al., 1996. Reed beds and constructed wetlands for wastewater treatment. ISBN 898920273 14. Curtis, T.P. et al., 1994. Light penetration in waste stabilization ponds. Water research 28, 1031-1038. Natural Systems for Wastewater Treatment in Warm Climate Regions Page 89 15. Davies-Colley, R.J et al., 1999. Inactivation of faecal indicator micro-organisms in waste stabilization ponds: Interaction of environmental factors with sunlight. Water Research 33 16. Davies-Colley, R.J et al., 2000. Towards a mechanistic understanding of pond disinfection. Wat. Sci. Tech. 42(10-11). 17. Decamp, O. and Warren, D., 1998. Bactivory in ciliates isolated from constructed wetlands (reed beds) used for WASTEWATER treatment. Water Research 32(7) 18. Diaz, O.A., et al., 1994. Solubility of inorganic phosphorus in stream water as influenced by pH and calcium concentration. Water Research, 28(8), 1755-1763 19. Donoso, C., 2013. Influence of heavy rain episode on removal efficiency in a three stage hybrid constructed wetlands system. UPC 20. EC, 2000. Directive 2000/60/EC of the European Parliament and of the Council of 23 October 2000 establishing a framework for Community action on the field of water policy. 21. EEC, 1991. Council Directive 91/271/EEC of 21 May 1991 concerning urban waste-water treatment. 22. Eger, P., 1992. Wetland treatment for trace metal removal from mine drainage: the importance of aerobic and anaerobic processes. IWA Publishing 23. Hammer, D., 1989. Constructed wetlands for wastewater treatment. Lewis Publishers 24. Gadd, G.M., 1990. Heavy metal accumulation by bacteria and other microorganims. Experientia 45, 834-840 25. Gale, N.L. and Wixsoc, B.G., 1979. Removal of heavy metals from industrial effluents by algae. Dev. Ind. Microbiol. 273 26. García, J. and Corzo, A., 2008. Depuración con humedales construidos, guía práctica de diseño, construcción y explotación de sistemas de humedales de subsuperficial. UPC Edicions 27. García, J. et al., 2005. Effect of key design parameters on the efficiency of horizontal subsurface flow constructed wetlands. Ecological Engineering 25(4), 405-418