Planted drying beds in the African context: state of knowledge and prospects

Although high rates of open defecation initially marked most African countries’ sanitation situations, multiple initiatives since the 2000s have reduced the severity of this problem and enabled the establishment of onsite sanitation systems. Because more than 55% of dwellings are not connected to a sewage system, onsite sanitation is usual (Ingallinella et al. 2002; Laminou Manzo et al. 2015; Nansubuga et al. 2016; Abubakar 2017). The management of sludge from these facilities, however, poses a new difficulty. The implementation of a system for emptying, transporting, and evacuating this sludge in a way to minimize population health hazards while still respecting the environment is still an issue. Fecal sludges (FSs) are dumped in nature on the outskirts of cities by emptying companies without being treated, exposing all the health and environmental dangers that this practice entails (Nansubuga et al. 2016; Bishoge 2021).

Most African countries currently face FS evacuation and management issues. Firstly because of the lack of logistics and the required organization to manage the large amounts of sludge produced, and secondly, the socio-economic context not allowing the use of the same technologies as in developing countries. Moreover, sludge management requires specialized technical understanding (Uggetti et al. 2010). Sludge treatment operations mainly involve two processes, namely (i) dewatering and (ii) stabilization (Kadlec & Knight 1996).

The goal of the dewatering process is to minimize sludge volume while increasing total solid matter (TSM) content in the finished sludge to save operating expenses. On the other hand, sludge stabilization tries to keep the biodegradable fraction of organic matter in the sludge stable by lowering the organic percentage of volatile matter (VM). Several sludge treatment technologies are based on these two unitary operations, i.e., sludge dewatering and stabilization. One of them, which is of major interest, is the planted drying bed (PDB). This technology is increasingly finding its place in the African context and deserves special attention. This literature review aims to discuss the current state of the use of PDBs in the treatment of FS in Africa, as well as additional research needs and perspectives. It is noted that in the scientific literature, there is no recent technical document that provides an update on the progress of this technology in Africa. Therefore, this work can serve as a summary of the different practices and techniques currently used in the field of PDBs for the treatment of FS. The next paragraphs will address the implementation and operation aspects of PDBs in Africa, the achieved treatment performance, observed challenges, and recommendations for future research in the field.

PDBs were invented in Europe in the 1980s to solve the inefficiency and clogging issues that plagued unplanted drying beds (Bittnann & Seidel 1967). This approach was developed by German researchers who decided to plant reeds (Phragmites australis) on unplanted drying beds to avoid clogging and the need for cleaning after each sludge feeding (Bittnann & Seidel 1967). The goal was to use the stem–rhizome–root system created by plants in the sludge layer to create water flow preferential channels. The technique was quickly embraced and applied in the 1990s in Germany (Hofmann 1990), Denmark (Nielsen 1990; Nielsen 2005), the United States (Kim & Cardenas 1990; Burgoon et al. 1997; Kim & Smith 1997), and France (Lesavre & Iwema 2002; Liénard et al. 2006; Troesch 2009; Vincent 2011) for activated sludge treatment. In the late 1990s and early 2000s, countries like Belgium (De Maeseneer 1997), the United Kingdom (Edwards et al. 2001), Poland (Hardej & Ozimek 2002; Obarska-Pempkowiak et al. 2003), Italy (Giraldi et al. 2009), Spain (Uggetti et al. 2009), and those in the South, where climatic conditions are more conducive to the usage of this technology, began to adopt it.

PDBs were initially developed to dewater and stabilize sludge from small activated sludge treatment plants in Europe and the United States (Kadlec & Knight 1996; Liénard & Payrastre 1996; Nielsen 2003). The process was then successfully adapted in other parts of the world for different types of sludge, including household FS.

PDBs are appropriate for tropical regions because seasonal climate fluctuations are minor and solar radiation is constant. Indeed, according to Edwards et al. (2001), in temperate climates, the performance of PDBs is better in summer when there is more sunlight than in winter (Kivaisi 2001; Nielsen 2005; Liénard et al. 2008; Dan et al. 2011). The use of PDBs in Africa dates back to 2005–2006, when authors such as Kengne et al. (2006), in partnership with the Swiss Federal Institute of Aquatic Science and Technology (Eawag), made some experimentations on the process. Kengne's research focused on selecting local macrophyte species (Echinochloa pyramidalis and Cyperus papyrus) as well as improving the system's treatment performance. Most PDB experiments in Africa were conducted on a pilot scale. This was the case for the experimentations conducted in Cameroon (Kengne et al. 2006, 2008, 2011; Ngoutane et al. 2012; Soh et al. 2014). However, several authors have conducted large-scale deployments in Senegal (Abiola 2009; Tine & Dodane 2009). Another example of large-scale implementation is the ‘Cambérène’ treatment plant in Dakar (Dodane et al. 2012). Some recent research was also conducted in Burkina Faso at the ‘International Institute for Water and Environmental Engineering (2iE)’. A pilot scale evaluation of the performance of indigenous species in PDBs has been done. In Benin, projects for the construction of sewage sludge treatment plants using PDBs are in progress.

The feeding of sludge to the beds is commonly done by batch system intermittently (Kouawa 2017) with heights up to 20 cm (Kadlec & Wallace 2008) or even 30 cm in some circumstances. Still, the preference is more evident for heights of 20 cm (Dodane & Mariska 2014). Due to the varying impacts of different sludge heights (obtained after feedings), which depend on the quality of this sludge, the reasoning based on applied loads expressed in kg TS/m²/year is more appropriate. Concerning the type of feeding, the batch feeding system is the most commonly used and recommended (Molle 2003). The pollutants in the sludge are subsequently removed using a combination of physical, chemical, and biological processes that result in the sludge getting dewatered, stabilized, and mineralized. Plants, microorganisms, and other physical variables play a role in biological treatment procedures in which various biochemical changes occur (Koottatep et al. 2005).

Several studies have revealed the increased use of plants in PDBs (Edwards et al. 2001; Stefanakis & Tsihrintzis 2012). Plants introduced in PDBs improve the sludge dewatering process and remove organic pollutants that act as nutrients due to their evapotranspiring properties and root network. Furthermore, the presence of roots encourages the expansion of microorganisms engaged in pollution elimination. Common reeds of the genus Phragmites sp. (Puigagut et al. 2007) and cattails of the genus Typha sp. are the most commonly used species in applying this technology (Stefanakis et al. 2014). In Africa, the dynamic is much more oriented toward promoting endogenous species (C. papyrus, E. pyramidalis, Echinochloa colona, P. australis, Phragmites vulgaris, Typha domingensis, Typha australis, Oryza longistaminata, Sporobolus pyramidalis, and Cyperus alopecuroides, Bambusa vulgaris) (Table 1). Some species, such as the fodder plant E. pyramidalis (Kengne et al. 2006; Abiola 2009), and others, such as the bamboo species B. vulgaris (Bakayoko 2020), are highly efficient and give good results. The literature also revealed that the species O. longistaminata, S. pyramidalis, and C. alopecuroides are not suitable for use in PDBs because all three species’ plants withered and died (Kouawa 2017).

The E. pyramidalis plant is currently the most effective in Africa. Purification yields were highest in beds with this plant: 97% for DM; 99% for total suspended solids (TSS); 99% for chemical oxygen demand (COD); 91% for NH₄⁺ and 97% for (Abiola 2009). In PDBs, E. pyramidalis has a significant rate of growth and multiplication. Some researchers noted that after one month, concentrations of over 100 individuals/m² could be achieved (Kengne et al. 2006). Its capabilities enable this plant to quickly colonize the bed's surface and build a root and stem network that cracks the sludge layers (Abiola 2009), enhancing the system's hydraulic performance, aeration of the upper layers through these fissures, and oxygen release from the roots. E. pyramidalis is also a robust plant, able to tolerate the water stress that is common in PDBs, as well as the high fluctuations in pollutant load, particularly salinity rate, which can be harmful to other plants very quickly (Kouawa 2017).

Comparative tests of Phragmites sp. and cattails of the genus Typha sp. for the treatment of agri-food sludge demonstrate that Phragmites sp. has a higher evapotranspiration (ET) rate (Wang et al. 2009). Furthermore, under tropical conditions, comparison tests between E. pyramidalis, P. australis, and T. domingensis reveal that E. pyramidalis has the best ET rates, with percolate percentages of 34, 38, and 47%, respectively, indicating that E. pyramidalis has the best ET rates (Sonko 2015). Comparative studies by Dodane et al. (2011) on E. pyramidalis, T. australis, and P. vulgaris also clearly highlighted the robustness of E. pyramidalis compared to the other two.

It should be mentioned, however, that other Echinochloa species, such as Echinochloa callopus and Echinochloa stagmina, may be more effective than E. pyramidalis. Still, no comparative research has been done on them. In addition to E. pyramidalis, Gueye et al. (2016) found that Echinochloa crus-galli, Paspalidium geminatum, and Paspalum vaginatum are all suitable candidates for FS treatment. However, compared to the other plants available, E. pyramidalis is still the finest, considering its numerous advantages (size, density of the root network, robustness). At a certain stage of the system's evolution, it becomes necessary to mow down the plants. Plant harvesting from PDBs refers to the process of removing and harvesting plants that have been grown in drying beds as part of a treatment system for sludge.

The harvested plants are typically removed from the beds once they have reached maturity, and they can be disposed of or used for various purposes. Depending on the type of plants used, they may be composted, used for bioenergy production, or even consumed as a food source (Gueye et al. 2016). Mowing down the plants is generally carried out during the cleaning of the beds, but it can be performed at other times as needed. For example, it is necessary to mow down the plants during insect attacks to allow young and more vigorous plants to take over. This activity is still done manually, but mechanized equipment is necessary for large stations. Plant harvesting should be done by cutting the plants at the surface of the beds without uprooting them. This avoids damaging the filtration matrix and allows the remaining rhizomes to continue functioning.

Plant harvesting from PDBs is an important aspect of the overall management of these systems, as it helps to maintain their effectiveness and prevent the buildup of excess biomass that can hinder treatment performance. It also provides an opportunity to recover valuable resources and reduce waste. Moreover, some authors have demonstrated that multiple harvests of plants increase biomass production (to be harvested and commercialized) and system processing performance (Rozema et al. 2016).

PDBs consist of several layers of granular media of increasing size from top to bottom, through which water percolates. The beds are typically composed of three layers (Table 2). The topmost layer is in contact with the mud and is typically made up of sand. Below the surface layer is a transition layer, which is usually composed of fine gravel. Finally, at the bottom of the bed is a layer of coarse gravel.

The top layer of the filter is common sand or fine gravel, with a uniformity coefficient larger than 3.5, to prevent hydraulic failure and clogging of the system. The total height of the layers varies from 30 to 90 cm (Paing & Voisin 2005; Abiola 2009), and the thickness of each layer, as well as the qualities of the material in each layer, varies from one author to the next. Some authors investigated the use of compost in place of sand and reported that compost increases rapid plant development during the acclimation period but decreases the quality of the leachate (Troesch 2009).

Aeration drains can be inserted between the layers to optimize the filter matrix's functionality. Comparative investigations of these aeration drains demonstrate that while their presence does not directly correlate with dewatering effectiveness, it maximizes plant and bacterial flora growth, resulting in better ET (Stefanakis & Tsihrintzis 2012). Authors usually used three layers of a granular medium.

While layer heights do not appear to substantially impact treatment efficiency, the particle size characteristics of the granular material that makes up each layer do (Stefanakis & Tsihrintzis 2012). Indeed, the greater the quality of the leachate, the smaller the particle size of the surface layer material; however, the risk of clogging also increases. This is the case of Abiola (2009), who used a very fine rolled dune sand of 0.5 mm as a surface layer and achieved one of the best purification results in the African context (Table 3). However, in West Africa, climatic conditions cause large ET rates, which significantly minimize the risk of clogging.

PDBs go through a feeding and resting cycle. These cycles can last from a few days to several weeks (Koottatep et al. 2005). The sludge is usually dumped on the bed's surface during feeding. Surface load calculated in the quantity of dry matter brought on a bed per unit of surface in one year (or kg TS/m²/year) is used to estimate the amount of sludge to be brought during the PDB's sludge feeding.

Breaking the volumes associated with retained loads into short cycles is more advantageous than using long cycles (faster dehydration of sludge) (Nielsen 2005). However, according to published research, the best time between feeds for high drying performance with a small number of beds is 11 days (Giraldi et al. 2009). The climatic environment of West African countries maximizes drying efficiency. As a result, these countries have shorter resting durations than countries with temperate climates for an equal amount of dry matter. In general, feeding cycles (number of feedings/total number of days in the cycle) in Africa vary around 3/7, 2/7, 1/7 (Kouawa 2017), and 1/15 (Abiola 2009; Kengne et al. 2009) (Table 3).

The climatic setting of West African countries, with a hot and dry propensity, makes them appropriate places for PDB implementation. The substantial association between climate and ET has been underlined in European studies (Stefanakis & Tsihrintzis 2011; Stefanakis et al. 2014). According to Giraldi et al. (2009), who also tested PDBs in Europe, summer results in a 40% improvement in dehydration performance. Water percolation due to gravity and ET due to climate are the two main mechanisms for sludge dewatering.

However, some African countries’ hot and dry climates allow engineers to maximize ET and, as a result, sludge dehydration. On the other hand, the authors found that aeration drains put in PDBs have no significant effect on boosting sludge dewatering. However, the oxygen supply promotes the development of microbial degradation processes in the filter bed, resulting in improved leachate quality (Stefanakis & Tsihrintzis 2012).

While percolation and ET are the keys mechanisms for dewatering (Cofie et al. 2006), sludge stabilization and mineralization are facilitated by aerobic bacteria adhered to the granular support and the roots of planted macrophytes’ rhizomes (Bialowiec et al. 2012; Gagnon et al. 2013; Brix 2017)

The fluctuation of loads affects the turbidity of the leachate and its physicochemical quality, according to many studies (Kengne et al. 2006; Vincent 2011). As a result, the higher the pollutant load, the higher the turbidity of the leachate. Kengne et al. (2006) and Vincent (2011) evaluated the effects of three loads on the performance of PDBs and found a positive correlation between loads and TSS and COD concentrations. However, the performance of PDBs in terms of purification remained superior to that of non-PDBs (Edwards et al. 2001). Indeed, Kengne et al. (2011) found average purification efficiencies of 95.8, 92.0, 95.4, 77.6, and 98.7% for TSS, total solids (TS), volatile solids (VS), NH₄⁺, and COD removal, respectively. Abiola (2009) achieved average purification efficiency of 97% for TS, 99% for TSS, 100% for TSS, 99% for COD, 91% for NH₄⁺, and 97% for ⁠, which is one of the most evocative cases. Despite these excellent results, direct disposal of leachate into nature is not suggested because of the high contaminants. The concentrations of leachate are often higher than the effluent discharge regulations in nature. Treatment options include lagoons or treatment at a second leachate PDB, as well as recirculation, which can be a more or less effective technique for improving leachate quality.

However, some researchers found that the leachate still has high coliform concentrations even with adding additional PDBs (Soh et al. 2014). The authors’ treatment scheme achieved compliance with WHO guideline thresholds for discharge or nonrestricted agricultural reuse at loading rates of 50, 100, and 150 mm/d for all monitored parameters, with the exception of nitrogen and fecal bioindicators. Studies have shown that treating FS in PDBs reduces helminth egg concentrations from a range of 600 to 6,000 helminth eggs/L of sludge to 170 eggs/g of dry matter (Ingallinella et al. 2002). In Africa, research shows that PDBs allow for the retention of helminth eggs in accumulated sludge and total absence in leachate (Kengne et al. 2009). In their study, Nzouebet et al. (2022) demonstrated that treatment with PDBs can lead to helminth egg elimination performance ranging from 82 to 100%.

PDBs have progressed well since their inception in Africa, but the growth of this technology has been far more empirical. There are still a lot of questions that need to be answered. The impact of variables such as water stress, salinity levels, climate (heat or cold tolerance), and illnesses (parasitic resistance) on plants like E. pyramidalis, which is now one of the most important species planted in Africa's drying beds, must be thoroughly investigated. This will help us understand the variables responsible for the plants’ mortality or wilting rates. The research should also focus on identifying new species that can be more powerful than those presently known, such as E. callopus and E. stagmina, or Panicum maximum, Echinochloa colona, E.crus-galli, Eleusine indica, Imperata cylindrical, P. geminatum and P. vaginatum which may be more effective than E. pyramidalis (Gueye et al. 2016). Research should continue to discover new and higher-performing native species and test the efficiency of various plant combination designs in the beds. Acrostichum aureum, for instance, has previously demonstrated its effectiveness in treating wastewater with high salinity levels, as is the case with sewage sludge (Sansanayuth et al. 1996). It is a species that not only thrives in mangroves but also in freshwater settings and salt marshes. It's a fast-growing, hardy plant (Akinwumi et al. 2022). This plant has also demonstrated significant phytoaccumulation performance in recent studies (Nguyen et al. 2021). Furthermore, species like Sesbania rostrata can be a strong choice for PDBs, due to their ability to withstand periods of both soil waterlogging and drought, as well as their capacity to tolerate high levels of salinity (Ramani et al. 1989; Naher et al. 2020).

From the standpoint of purification performance, it would be possible to achieve improved purification performances by including the parameter of leachate residence time at the bottom of the beds before evacuation in the design and dimensioning parameters of the PDBs. It would be helpful to focus research on this direction. Some authors are already conducting studies in this direction by testing combinations of PDBs in series (Soh et al. 2014) or by combining PDBs with maturation ponds in order to refine the treatment of leachates (Soh et al. 2022). In the same dynamic and with the aim of improving the quality of percolates, the combination of PDBs with wastewater treatment processes such as the use of floating macrophytes (Mahunon et al. 2018; Akowanou et al. 2023), microalgae (Liady et al. 2022), or activated carbon (Daouda et al. 2021) would be a promising research avenue.

The implementation of PDBs in Africa has made significant progress since their introduction to African sanitation. Among the various solutions adopted by Africa, particularly in sanitation, PDBs for the treatment of FS have proven to be an excellent alternative to conventional approaches. Therefore, they deserve the attention of the scientific community and leaders. However, there is still much room for improvement, especially regarding scaling up and adequate sizing considerations, taking into account population growth over time.

All relevant data are included in the paper or its Supplementary Information.

The authors declare there is no conflict.