Next Article in Journal
Spatiotemporal Characterization and Analysis of River Morphology Using Long-Term Landsat Imagery and Stream Power
Previous Article in Journal
Sequential Analysis of Phosphorus Compounds Contained in the Substrates and the Digestate
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:

Multi-Soil-Layering, the Emerging Technology for Wastewater Treatment: Review, Bibliometric Analysis, and Future Directions

National Center for Studies and Research on Water and Energy (CNEREE), Cadi Ayyad University, P.O. Box 511, Marrakech 40000, Morocco
Laboratory of Water, Biodiversity and Climate Change, Faculty of Sciences Semlalia, P.O. Box 2390, Marrakech 40000, Morocco
Faculty of Life and Environmental Science, Shimane University, Matsue 690-0823, Japan
Author to whom correspondence should be addressed.
Water 2022, 14(22), 3653;
Received: 23 October 2022 / Revised: 6 November 2022 / Accepted: 8 November 2022 / Published: 13 November 2022
(This article belongs to the Special Issue New Insights into Wastewater Reclamation and Reuse)


Due to its unique structure and excellent purification efficiency (e.g., 98% for organic matter and between 94 and 100% for nutrients), multi-soil-layering (MSL) has emerged as an efficient eco-friendly solution for wastewater treatment and environmental protection. Through infiltration-percolation, this soil-based technology allows pollutants to move from the MSL upper layers to the outlet while maintaining direct contact with its media, which helps in their removal via a variety of physical and biochemical mechanisms. This paper attempts to comprehensively evaluate the application of MSL technology and investigate its progress and efficacy since its emergence. Thus, it will attempt via a bibliometric analysis using the Web of Science database (from 1993 to 1 June 2022) related to MSL technology, to give a clear picture of the number of publications (70 studies), the most active academics, and countries (China with 27 studies), as well as collaborations and related topics. Furthermore, through hybrid combinations, pollutant removal processes, MSL effective media, and the key efficiency parameters, this paper review will seek to provide an overview of research that has developed and examined MSL since its inception. On the other hand, the current review will evaluate the modeling approaches used to explore MSL behavior in terms of pollutant removal and simulation of its performance (R2 > 90%). However, despite the increase in MSL publications in the past years (e.g., 13 studies in 2021), many studies are still needed to fill the knowledge gaps and urging challenges regarding this emerging technology. Thus, recommendations on improving the stability and sustainability of MSLs are highlighted.

1. Introduction

Adequate sewers and treatment plants are still lacking in many parts of the world, especially in the rural areas of economically developing countries where wastewater treatment infrastructure is limited or inadequate [1,2]. Furthermore, implementation and maintenance of centralized wastewater treatment systems in these areas are often challenging and costly [3]. On the other hand, inappropriate sanitation systems contribute to a variety of parasitic infections and negatively affect population health as well as socio-economic development [3]. Since wastewater management is a major issue and given the increase in the amount of wastewater discharged in rural areas, researchers have begun to consider more environmentally friendly, cost-effective, and long-term treatment solutions [4].
In this regard, multi-soil-layering (MSL) is an innovative and eco-friendly wastewater treatment system [5]. MSL is based on the infiltration-percolation of wastewater and involves alternating layers of soil mixture with permeable layers (e.g., zeolite and gravel). It offers many benefits, including small-scale implementation, low cost, ease of maintenance, and the possibility of reusing its effluents in the agricultural field [6,7]. Furthermore, several studies have acknowledged its effectiveness in removing pollutants from wastewater through combined physical, chemical, and biological mechanisms [7,8].
Although, recent studies have been concerned with the bacterial community in the MSL system and its sanitary efficiency [5,6,9,10,11]. Thus, some concerns about the MSL’s performance are still unclear [12,13]. For instance, the removal of fecal indicator bacteria has moved from 1.01 to 1.28 log units using a single-stage MSL system [7,14]. However, given the necessity to reuse the MSL effluent in agriculture, this percentage still falls short of the standard level [6]. Therefore, the vital challenge with MSL is removing fecal indicator bacteria and pathogens, whose level in treated water is a sign of good sanitary efficiency.
The current work sheds light on a number of subjects that MSL’s earlier research did not address. For instance, most of the studies are still locked in laboratories and at the experimental level [15,16,17], whereas the full-scale experiment still has a weakness [18]. Regarding MSL efficiency, although the removal of organic matter and phosphorus is very promising [19,20], total nitrogen abatement efficiency has not yet achieved this level [13,21]. Moreover, despite using a two-stage MSL system with a vertical flow, the high removal rate of coliform bacteria recorded did not exceed 3.15 log units [7], requiring the development of additional methods to improve MSL sanitary efficiency. Similarly, there is a weakness in research on viruses, pathogens, emerging pollutants, and their removal mechanisms [14,22]. As for combinations, several hybrid configurations are still under experimentation [23,24,25,26,27]. However, in this literature review, it was observed that there is a lack of studies investigating two-stage MSL with vertical and horizontal flow [18]. Regarding modeling approaches, although their results were encouraging [6,10,24], the application of other powerful methods and hybrid models may help to investigate pollutant removal in the MSL system [28,29].
Previous studies have mainly focused on the removal of organic matter and nutrients [30,31,32], whereas recent research has focused more on coliform bacteria and microbial activity [5,7,9,11,33,34]. In this regard, this paper summarized and discussed MSL performances and highlighted the most studied pollutants and those receiving the least attention. In addition, this review tries to describe the pollutant removal mechanisms based on the latest studies, especially with regard to coliform bacteria, which represents one of the shortcomings of the MSL system. In the same sense, both prior and recent studies agree that aeration, hydraulic loading rate (HLR), and clogging are the key MSL parameters [10,12,16,35,36]. Thus, the current paper attempts to examine their relationships in the MSL field using keyword co-occurrence analysis. Regarding MSL application, most prior studies were carried out at the laboratory or pilot scale [6,15,21,37]. Nowadays, researchers have turned to combining MSL with other treatment technologies, such as constructed wetlands and trickling filters, at full scale [20,23]. Thus, this paper compares and evaluates their efficacy to determine which hybrid configuration is most effective in removing pollutants. Regarding MSL modeling, this topic is still emerging, and the applied nonlinear methods have accurately simulated MSL performance [6,10,16]. Thus, this paper attempts to quantify and discuss the types of applied models, compare their accuracies, and investigate the most targeted parameters.
On the other hand, for the first time, this paper attempts to investigate and explore the evolution of the MSL technology through bibliometric analysis, which has expanded in recent years due to the urgent demand for these types of studies based on quantitative methodologies, especially in the area of wastewater treatment [38]. In addition, bibliometric analysis has evolved into a useful tool for assessing the influence of scientific publications in a given research field and has become extremely important in assessing the quality of studies [39]. Subsequently, a comprehensive investigation and a detailed description of various aspects of MSL technology (e.g., efficiency, design, operating conditions, modeling approaches, etc.) were carried out to offer an overview and identify knowledge gaps and potential future challenges.
Based on the literature and limitations, the main focus of this bibliometric and review paper is to find out the effective working of multi-soil-layering eco-technology, considering the key efficiency parameters, modeling approaches, and key knowledge gaps.
The main objectives of this paper are:
to conduct a bibliometric analysis in the field of MSL technology;
to provide a comprehensive summary of MSL aspects (e.g., performance, removal mechanisms, etc.);
to issue a comparative profile of the MSL with other eco-friendly technologies;
to evaluate models applied to simulate MSL performance;
to highlight the MSL challenges and provide a road map for future research.

2. Materials and Methods

To define the MSL field, the first approach used in this study consists of analyzing the visibility of scientific publications on the MSL system using the library of Web of Science (WoS) (published by Thomson Reuters) between 1993 and 1 June 2022. This catalog of bibliographic references provides access to a wide range of scientific databases and information services and contributes to the advancement of bibliometric analysis. Due to its reliance on the quantitative and statistical method by which scientific production is analyzed and characterized, authors’ productivity determined, and scientific production evaluated, bibliometric analysis is currently regarded as a self-contained scientific research approach [40,41]. Subsequently, we have applied a progressive research process to determine the relevant keywords that encompass the publications of this technology, while eliminating those that took us away from the term “multi-soil-layering”.
On the other hand, the most commonly used words in the MSL literature were studied. Thus, links between pollutant removal and these extracted words were discussed to understand the MSL system’s treatment behavior. Therefore, we have chosen the following keywords: (“multi-soil-layering”) or (“vertical multilevel soil”) or (“multi-soil layer” and “MSL”) or (“multiple-soil-layer” and “wastewater” or “multi-media-layering” and “mixture”). Thus, only English-language documents were considered. Regarding co-occurrence analysis, bibliometric maps were created using VOSviewer software (version 1.6.17). In addition, R software (version 3.5.2) was used to show statistics related to MSL publications (e.g., researchers, countries, journals) and collaborative networks between countries in the MSL field.

3. Results

3.1. Bibliometric Analysis

3.1.1. MSL Published Papers

The application of MSL eco-technology to rural sanitation has been studied worldwide. In this sense, a total of 70 publications were found in the WoS database from 1993 to 1 June 2022. First of all, only 21 papers are freely accessible (open access), and Figure 1a presents the annual MSL publications. Scholars and academics were initially uninterested in this technology when it was first introduced in the early 1990s, until two peer-reviewed scientific papers were published in 2002. Furthermore, the number of publications has increased considerably in recent years, reaching a high of 13 in 2021 (Figure 1a). This result demonstrates an increasing interest in investigating the MSL over the years.
Figure 1b displays data on the number of papers published in each journal. According to the WoS database, seven papers were published as proceeding papers. Thus, Soil Science and Plant Nutrition was the scientific journal that published the highest number of documents regarding this technology, with 11 articles, followed by the Ecological Engineering and Bioresource Technology journals with seven and four articles, respectively. Refereed journals that have published two articles in this regard were also mentioned, among them Environmental Technology.
Although MSL technology is still being increasingly researched in different countries around the world, China is considered the most significant contributor to the development of this technology with 27 studies, followed by Japan with 22 studies. It was also noted that other countries outside of Asia are interested in studying this technology, as Morocco ranked third in terms of the number of published documents with 12 research papers (Figure 1c). In the same context, related to the authors’ publications, Figure 1d displays the number of papers for each author during the period (1993–2022). As it can be observed, Masunaga T, Wakatsuki T, and Sato K are the researchers who have contributed the most to the development of MSL technology, with a total of 18, 16, and 10 papers, respectively. Furthermore, these authors have jointly published numerous articles on different aspects of the MSL system since 2005. Thus, a vivid illustration of this collaboration is the newly released paper on MSL’s long-term efficiency [1].

3.1.2. MSL World Collaboration

The countries that produce scientific papers about MSL technology are limited to certain regions and are distributed unevenly around the world, as the geographical distribution of scientific publications shows that they are mostly concentrated in East Asia, North America, and North Africa (Figure 2a). In addition, nearly 70% of all MSL publications come from Asia. As it can be observed in Figure 2b, the collaborative analysis of countries and institutes highlights the primacy of Japan (total collaboration = 15). The cause is probably twofold: Japan was the first country to study the MSL system, and the second reason is due to the strong presence of Japanese researchers in collaborative projects. As a result, even some neighboring countries, such as China and Thailand, collaborate with Japan but not with each other. Additionally, five regions revolve around Japan, represented by countries from Asia, North Africa, North America, and Europe. For instance, an important cooperative network with Thailand is shown by the seven contributions in green (Figure 2). The Japanese researchers also shared their knowledge about MSL technology and collaborated with Morocco through four publications (orange) and the USA through two publications (purple). China, as the first producer of scientific articles on the MSL system, has also invested in collaborations (total collaborations = 14), alongside Canada and the USA with ten publications.
Recently, the collaborations with China are extensive and have also reached countries in Africa (Tanzania and Zimbabwe) and Europe (Germany) each with a single collaboration. In addition, Morocco, which ranked as the third country in terms of the number of articles published on MSL technology, has collaborated nearly seven times with four countries scattered all over the world, and this is for the entire period from 2014 to 2021 (Figure 2e).

3.1.3. Co-Occurrences Analysis

Through the search for the most relevant co-occurring items, defined by their relative frequency, we try to make the most repeated items appear. Thus, a bibliometric map of co-occurring terms will be useful in determining which ones best characterize the MSL field. Using density visualization in VOSviewer software (Figure 3), the extracted items shown in the map are exhibited by a label, where the color of each circle indicates its frequency of co-occurrence (density). Colors from the rainbow palette were used (red, orange, yellow, green, and blue); the closer the circles are to red, the more important the element is within the network map. Conversely, the closer the color of the circle is to green or blue, the lower the occurrence of items and the lower their frequency [42].
It was feasible to verify that the hotspots of multi-soil-layering systems, efficiency, performance, nitrogen removal, phosphorus, organic matter, mechanisms, constructed wetlands, and wastewater treatment were the frequently prominent items. It is obvious from the bibliometric map that the frequency of co-occurrence items suggests that researchers are still working on developing and upgrading MSL technology by looking at the removal of conventional contaminants that are known to affect water quality (organic matter and nutrients). However, there is a weakness and an absence of words related to pathogens and coliform removal as a critical indicator of water quality and their accompanying removal mechanisms. Constructed wetlands’ inclusion in this bibliometric map of co-occurrence analysis might be due to researchers identifying them as an alternative technology to the MSL on the one side or comparing their efficacy to the MSL system on the other. Furthermore, recent research has integrated these two systems into a combined system to investigate their efficacy in treating wastewater.

3.2. Main Pollutant Removal Mechanisms in MSL

The operation of the MSL system is based on infiltration-percolation processes, with the soil mixture serving as the main purification media [43]. The MSL configuration design and method of arranging its components play (Figure 4) a critical role in the distribution of effluent to the MSL layer’s surface and help to reduce the risk of clogging [32]. In addition, the soil is a vital ecosystem comprising water, air, organic matter (e.g., animals, plants, and microbes), and inorganic material (e.g., minerals) [44]. It may play a significant role in pollution control by attracting contaminants in wastewater that travels through it, due to its absorption capacities.

3.2.1. Suspended Solids

Suspended solid (SS) removal was mostly accomplished by filtration as a physical process at the MSL upper layers, and their buildup in these surface layers increases the hydraulic residence time of wastewater within the MSL system [37]. Furthermore, the size of the SS and the MSL media porosity are important factors in SS removal. According to Ho and Wang [30], under the HLR of 1000 L/m2/day, permeable layer (PL) materials with pore sizes between 3 and 6 mm had an SS removal rate of more than 80%. Despite the MSL system’s ability to remove fine particles, Ho and Wang [30] recommended that the HLR be increased to decrease the risk of clogging. However, while investigating the performance of MSL systems under various HLRs.
Masunaga et al. [32] discovered that the removal of SS is not appreciably affected by changes in the HLR. Furthermore, to reduce the risk of clogging, it may be recommended that the MSL system be given a periodic rest time and that the removal of SS discharges into the MSL be preceded by a pre-treatment of wastewater [12].

3.2.2. Organic Matter

Regarding the organic matter, the aerobic and anaerobic biological decomposition of contaminants is based on the microbial biomass that makes up the MSL biofilm [9]. When wastewater is released into the MSL system (Figure 5), organic contaminants are first removed from MSL systems by physicochemical absorption by soil mixture particles and the surface area of the PL, and subsequently broken down and metabolized by heterotrophic aerobic microorganisms [9,12,19]. Thus, its decomposition leads to the emission of carbon dioxide (CO2) and methane (CH4) [45]. Furthermore, organic matter is removed mainly at the upper layers of the MSL system, where aeration is more favorable and provides a suitable environment for aerobic heterotrophic bacteria [9,12,19]. Song et al. [5] and Zhou et al. [9], in turn, found that temperature has a role in the removal of organic matter in the MSL system, with low temperatures being unfavorable for bacterial growth and metabolism.

3.2.3. Fecal Indicator Bacteria and Pathogens

Regarding microorganism removal efficiencies, the mechanisms involved in fecal bacteria indicator (FIB) and pathogen removal in the MSL system are filtration, adsorption, predation, inactivation, and die-off [7,37]. Thus, both permeable layers and soil mixture layers contribute to reducing their level through filtration (Figure 5), while adsorption is expected to occur in particular at the soil mixture layers [37]. As for microbial degradation, the influence of inhibitory compounds released by other bacteria in soil [46] or predatory populations (protozoa, worms), as reported by Kadam et al. [47], might play a significant role. Furthermore, the MSL system is an ecosystem that includes many species of bacteria [9,17,48] competing to survive and gain an advantage in terms of space and resources, including organic carbon and nutrients (nitrogen and phosphorus). Therefore, interactions within MSL and between microbial species may lead to decreasing these advantages for coliforms, which will negatively affect their level in the MSL system. In the same vein, Latrach et al. [7] claim that natural indicator bacteria die-off is induced by several inactivation mechanisms, including exposure to physicochemical stressors such as pH, oxygen, and temperature. Thus, the critical role of pH in removing FIB has been confirmed recently by Sbahi et al. [11], who also stated that pH action is reinforced by the denitrification mechanism and alkaline properties of soil, while iron metal contributes to decreasing fecal pollution level as an antibacterial element (Figure 5).

3.2.4. Nitrogen

The soil mixture layer and PL alternately contribute to the removal of nitrogen as wastewater passes within the MSL system. Thus, as reported by several authors [12,35,49], PL filled with zeolite can adsorb ammonium (NH4+) due to its strong cation exchange capacity. In the MSL system, after the transformation of total Kjeldahl nitrogen (TKN) to NH4+ through ammonification, the latter oxidizes in the combined presence of oxygen and nitrifying bacteria into nitrites (NO2) which in turn are transformed into nitrates (NO3) in the same conditions (Figure 6).
This biological process (nitrification) is localized in the MSL upper layer, where the number of nitrifying bacteria is higher due to an increase in oxygen and nutrient availability [50]. Furthermore, these wastewaters flow due to gravity, percolate through the PL, and then drain into the soil mixture layers (anaerobic zone), where they will be reduced in the presence of both organic carbon and denitrifying bacteria to nitrogen gas (N2, N2O) [7]. Soil organic matter, sawdust, and charcoal as carbon sources help to promote the denitrification process, which is controlled by degradable organic matter, NO3, and oxygen [49].
Furthermore, Epsilonbacteraeotra was discovered to be a crucial species in nitrogen removal by promoting denitrification at the oligotrophic level, whereas organic might improve the energy metabolism of microorganisms and enzyme activity (e.g., kinase, glycogen phosphorylase), strengthening the denitrification process [9]. In addition to adsorption, nitrification and denitrification are the main mechanisms of nitrogen removal in the MSL system. Thus, the coexistence of aerobic/anaerobic environments is one of the most essential factors influencing the effectiveness of the MSL system in removing nitrogen [7,12,51].

3.2.5. Phosphorus

Regarding phosphorus removal, it can be adsorbed on Fe and Al hydroxide contained in soil and zeolite [12,49], whereas ferrous ion (Fe2+) and ferric ions (Fe3+), as well as other cations present in the soil mixture, can co-precipitate with this pollutant [9]. Indeed, during the percolation of wastewater within the MSL layers, the iron added in the soil mixture layers is transformed into Fe2+, which is then transported to the PL and oxidized to Fe3+, allowing the removal of phosphorus (Figure 7). Zhou et al. [9] also claimed that soil mixture fillers with high surface areas, such as charcoal, have many micropores that contribute to phosphorus adsorption, while Sato et al. [19] indicated that the flow rate and permeability of the soil mixture had a significant impact on phosphorus removal efficiency in the MSL system. Regarding PL, it assumed greater responsibility for its removal than the soil mixture media in the case of high HLR [52] (Figure 7).

3.3. Key Efficiency Parameters

The effectiveness of the MSL system, like the majority of soil-based wastewater treatment systems, is influenced by a variety of parameters.

3.3.1. Temperature

For instance, Sbahi et al. [6,11] have reported that temperature fluctuations are among the most important parameters that influence coliform removal in the MSL system. Similarly, Guan et al. [36] stated that temperature is a limiting factor for bacterial growth in the MSL system and that this system can remove more than 95% of the five-day biochemical oxygen demand (BOD5) in the temperature range (25–35 °C) in a humid environment. Zhou et al. [9], in turn, have reported that biological processes within the MSL system, including the removal of nitrogen and organic matter, are sensitive to temperature. However, Latrach et al. [7] have claimed that seasonal temperature fluctuations did not influence MSL performance for both physicochemical and bacterial indicators under arid conditions.

3.3.2. Aeration

Regarding aeration, Song et al. [20] found that, while bottom submersion had the most negative effect, aeration was the most dominant beneficial factor for pollutant removal. Sato et al. [1] reported that anaerobic conditions have decreased the presence of ferric hydroxides, and as a result, phosphorus adsorption has decreased. Luanmanee et al. [35] also reported that aeration is an important factor in the operation of the MSL and recommended varying the duration of aeration depending on the quality of the wastewater and the components of the system. Zein et al. [53] also observed that aeration contributes to the removal of nutrients, oil, and grease in the MSL system.
However, excessive aeration could intensify nitrification and harm microorganisms’ metabolism [54], which may, in turn, inhibit the denitrification process and consequently reduce nitrogen removal [55]. In the same context, Luanmanee et al. [31] have declared that continuous aeration was successful in increasing the removal of BOD5 and total phosphorus (TP) inversely to total nitrogen (TN) removal (dropped to 31%). As a result, regulating the aeration of the system is necessary for maintaining the capacity of the MSL to remove organic matter and nutrients and to prevent clogging. In addition, Sato et al. [19] proposed that improving the MSL’s upper layers’ aeration and addition of organic matter in its lower layers of soil mixture were effective in simultaneously removing organic matter and nutrients.

3.3.3. pH

On the other hand, the effectiveness of the MSL system is related to the pH value of influent wastewater. For instance, Sbahi et al. [6,11] have reported that the pH of the influent was the most important variable that contributed to coliform removal in the MSL system. Similarly, Latrach et al. [7,37] reported that pH value is a critical factor controlling the survival of bacteria in alkaline water and would aid in their inactivation in the MSL system. In the same context, Song et al. [5] stated that pH could affect bacterial abundance and the nitrification process. In addition, they found that an acidic influent (pH = 3) was suitable for removing sulfamethoxazole through increasing its sorption affinity with the micropores of MSL media.
Regarding nutrient removal, Luanmanee et al. [35] found a significant (p < 0.05) positive relationship between TN and TP removal and the pH value of MSL effluent, which is also related to MSL aeration. Furthermore, they recommended that the pH of the MSL system’s effluent be kept between 6.5 and 7.0 to maintain its effectiveness in wastewater treatment. On the other hand, changes in the pH of the MSL effluents may be utilized to regulate the aeration rate and its duration in the MSL system [35,56] and consequently enhance the biological nitrification/denitrification processes.

3.3.4. Hydraulic Loading Rate and Clogging

Regarding the HLR and flow pathways, Sato et al. [1] have noticed that shortcut flow can cause a decrease in the decomposition of organic matter and the capacity for oxidation-reduction potential (ORP) of treated water. Guo et al. [57] and Sato et al. [19] have suggested that flow rate is a critical operational parameter for organic matter and nutrient removal and that its regulation can prevent clogging [58]. Guan et al. [36] have stated, in turn, that HLR strongly affects the distribution of anaerobic and aerobic zones by changing the depth of water. Guo et al. [57,59] concluded that the MSL system demonstrated strong adaptability with the variation of HLR.
However, Masunaga et al. [60] observed that HLR contributes to MSL clogging, especially at 4000 L/m2/day. In the same sense, Guan et al. [61] reported that HLR and hydraulic residence time are related factors and concluded that the residence time of flow inside the MSL system was highly correlated to HLR. As a result, a high inflow rate would favor short-circuiting flow, increase the volume of dead space, and consequently decrease MSL efficiency.
Regarding clogging, this factor has already been discussed by Guan et al. [36] as a crucial parameter in MSL efficiency. Thus, Figure 8 presents the bibliometric map when the key efficiency parameters “HLR (Figure 8a)”, “Clogging (Figure 8b)”, and “aeration (Figure 8c)” are selected. Through these network maps, the selected terms are related to each other and also affect the efficiency of the system through some of the relevant links (relationships with “multi-soil-layering system”, “efficiency”, and “performance”). In addition, they are directly linked to the mechanisms responsible for removing pollutants in the MSL system. Thus, selecting the item “hydraulic loading rate” highlighted some related terms, including “organic matter”, “nitrogen removal”, “denitrification”, and “performance”. In this context, increasing the HLR inversely reduces the contact time of nitrogen within the MSL media and consequently reduces its removal through limiting the denitrification process. In general, decreasing the HLR increases the removal of pollutants in the MSL system.

3.4. Filter Media and Structure

The lack of sufficient knowledge of the MSL system can lead to environmental effects on public health [36]. Sato et al. [19] stated that the effect of used materials and their structure within the MSL on the removal efficiency is not entirely understood. Thus, for the effective operation of the MSL system, researchers have extensively tested different kinds of materials. The soil mixture consists of soil (around 70%), while supplementary purification materials (iron, jute, sawdust, charcoal, and activated charcoal) depend on the type of wastewater and differ according to the research purpose [12,60].
The PL material consisted mainly of zeolite due to its high capacity of adsorption and affinity with NH4+ [31,60,62,63], whereas many studies confirmed the high purification capabilities of other materials such as gravel, perlite, and pozzolan [1,11,15,33]. In the same context, Maeng et al. [22] have reported that the introduction of zeolite/slag into the MSL system could improve the removal of traces of organic pollutants as well as the growth of bacterial biomass, resulting in better dissolved organic matter reduction via biodegradation. Song et al. [50], while treating turtle aquaculture effluent by an MSL system, found that adding sludge (20%) to the MSL could improve the NH4+ removal efficiency by promoting the number of nitrifying bacteria. Ait-Hmane et al. [64] have discovered, in turn, that MSL combined with activated carbon as an adsorbent was a suitable alternative for purifying wastewater from olive mills by improving the adsorption process. Wang et al. [65] have found that the purifying capacity of sludge-based biochar material outperforms that of charcoal in eliminating organic matter and nutrients.
Regarding the number of layers and configuration design, Sy et al. [66] discovered that the more MSL layers, the greater the decrease of contaminants, while Sato et al. [21] observed that six layers were found to be required for a high removal rate of organic matter. Moreover, Sato et al. [58] also indicated that reducing the width of soil-mixed blocks helped to improve the dispersion of water flow inside the MSL system. Chen et al. [43] reported that a larger soil mixture surface area increases the removal of SS, organic matter, and TP due to increased interaction and contact between the soil mixture layers and wastewater. Similar results were reported by Sato et al. [1]. In the same context, Wei and Wu [49] and Guan et al. [36] have stated that effluent recirculation is recommended to remove even more nitrogen in the MSL system.
Regarding soil mixture fillers, Xiao et al. [67] reported that combining granular Fe0 with an iron sulfide mineral improves the efficiency of conventional Fe0 systems, including MSL technology. In addition, sulfide minerals increase Fe0 dissolution by lowering the pH of the system, resulting in the subsequent precipitation of hydroxides coupled with the co-precipitation of contaminants. Chen and Pat [68] observed that iron was the main factor that improved the capability of the MSL in phosphorus removal (more than 83%). The same authors reported that the addition of oyster shells successfully eliminated up to 99% of the iron released from the soil mixture in the effluent after the reaction with phosphorus.
On the other hand, there are recent studies that have used new materials to improve the effectiveness of the system. For instance, Song et al. [48] reported that using polybutylene succinate (PBS) in soil mixture was favorable for species involved in the denitrification process. Hong et al. [13] reported that PBS had some advantages over sawdust as it is a pure carbon source and does not bring nitrogen and phosphorus to the MSL system. Additionally, Liu et al. [69] observed that PBS releases organic matter at a slower rate, which consequently increases the operation time of the MSL system. Zhou et al. [9], adding a blended organic carbon source (BCS) to the MSL system, have found that BCS increases denitrification gene abundance (nirS and nosZ). Guo et al. [57] reported that using a biosurfactant in the MSL was a promising option for enhancing ammonia removal and biological activity. Pattnaik et al. [70] showed that adding a sucrose solution to untreated effluent improved the removal of nutrients in the MSL system.
In general, the MSL system’s structure and component arrangement optimize wastewater dispersion and distribution while lowering the risk of clogging. In the permeable zone, zeolite is the most commonly employed filter media because of its high purifying power through adsorption and ion-exchange capacity, as well as its benefit to microorganisms. However, in contrast to sites marked by ancient volcanic activity [71], this material is not abundant in many parts of the world. Therefore, a low-cost and easily accessible alternative natural filter media, such as gravel, that has been proven to be effective is also a suitable choice.
On the other hand, the number of layers in MSL is an important factor in the system’s efficiency. In this regard, the more significant it is, the longer pollutants will be retained and the system’s efficiency will improve. In the same vein, the MSL’s efficiency rises in lockstep with the increase in the pollutants’ contact surface with the soil mixture blocks. Furthermore, adding an extra carbon source to the composition of these blocks might help the bacterial population grow and boost its action in the breakdown of organic matter and nitrogen. On the other hand, iron remains a critical parameter in the removal of phosphorus from wastewater, which may be enhanced further with the addition of sulfide minerals.

3.5. MSL Typology and Performance

3.5.1. Single-Stage MSL

The use of eco-friendly technologies, especially in rural areas without suitable purification facilities, is vital for reducing pollution from wastewater released into the natural environment. MSL is a natural technology developed for lowering pollutant levels in discharged effluents. Many researchers have tested the efficacy of this system in the lab and on a pilot scale [7,48]. However, its application at full scale is limited and focused mostly on domestic wastewater purification [1]. Several studies have looked into the effectiveness of the MSL system by experimenting with different configuration designs and purification materials [30,33,72].
Figure 9 depicts various types of wastewater treated by the MSL system at various scales (lab, pilot, and full scale), as well as the highest removal efficiency recorded for each water quality indicator. Across all MSL scales, real and synthetic rural domestic wastewater remain the most investigated kinds of water [6,17]. In addition, the MSL’s performance in treating other types of wastewater, such as textile and livestock effluent, is also being investigated [66,73].
Recently, on a laboratory scale, MSL has been demonstrated to be successful in removing a variety of contaminants, such as sulfamethoxazole in poultry wastewater and cyanotoxins in contaminated freshwater [5,15]. In terms of removal efficiency, MSL has demonstrated remarkable results in the removal of a wide range of pollutants. Regarding SS, in several investigations, its removal has exceeded 97%, both in the lab and at full scale [17,43,53]. In the case of organic matter, the chemical oxygen demand (COD) parameter was more thoroughly investigated than BOD5 (Figure 9). Thus, Sato et al. [19] have found that MSL removed almost 98% of BOD5 from domestic wastewater under an HLR of 1000 L/m2/day. Similar results were found for COD removal at HLRs of 400 and 500 L/m2/day, respectively [10,13]. However, Attanandana et al. [74] have found that MSL removes an average of 57% under a hydraulic head of 850 L/m2/day. This might be because the MSL system is still in its early stages, and that the findings of subsequent research have enhanced its efficiency. Concerning nitrogen removal, in the study conducted by Song et al. [20], the removal of NH4+ from domestic wastewater has reached 100%. Close results were recorded by Guo et al. [57] when treating swine wastewater, with a removal rate going up to 94% for TN under an HLR of 160 L/m2/day. In terms of TP, many studies have reached a removal rate of 100% [5,10,13].
Although these findings are encouraging, more research on the removal of FIB from wastewater is needed to increase the MSL’s sanitary efficiency, particularly at the full-scale level (Figure 9). Overall, the MSL system was able to remove 2.26 log units from domestic wastewater [33], whereas Sbahi et al. [11] discovered that the MSL system could remove 1.62 log units from rural domestic wastewater.
In terms of pH, the literature analysis revealed that it is about 8 in the MSL effluent (Figure 9). This suggests that the MSL system’s chemical activity has tended to be alkaline, which aids in the removal of contaminants [6,53,75]. Within the same context, Latrach et al. [7] and Masunaga et al. [32] observed that pollutant removal seems to not be largely influenced by seasonality. However, Sbahi et al. [6,11] found that the removal efficiency of the MSL system was sensitive to seasonal variations in semi-arid climates.

3.5.2. Combined Systems

It has been noted that the combination of the MSL with other treatment technologies in sanitation is not frequent around the world. Among the 70 papers considered, 13 studies investigated the combination of the MSL with other treatment plants. In order to efficiently treat different types of wastewater, the researchers have attempted to combine the MSL system with different other technologies, such as sand filters, trickling filters, constructed wetlands (CW), as well as MSL on a two-stage level. Table 1 shows a summary of the combined typologies’ performance. In addition, Figure 10 shows a schematization of these configurations based on the literature review.
  • Two-stage MSL
In this combination of two MSL systems, domestic wastewater flows under the influence of gravity (Figure 10a). Thus, it has shown a high potential for eliminating SS (97%), organic matter (≥91%), and nutrients (≥95%) from domestic wastewater in a small rural community in Morocco [7]. In addition, this configuration was shown to be efficient in fecal coliform bacteria removal (>3 log units) (Table 1). For the treatment of polluted river water, Wei and Wu [49] also developed a hybrid two-stage MSL system with vertical flow (Figure 10b). As a result, the authors discovered that a 50% bypass combined with the addition of an organic carbon source was able to accomplish the full nitrification process (NH4+removal = 99%). Furthermore, the study revealed a significant removal efficiency of COD (>70%) and TP (82%) (Table 1). Thus, in contrast to the system proposed by Latrach et al. [7], which uses gravel as a purification material, this configuration uses zeolite and ceramsite as purifying materials. In addition, the first unit had a ryegrass-planted topsoil layer.
  • MSL with constructed wetland
Gravity is used to move wastewater in these configurations. Furthermore, some research coupled MSL and CW on a two-stage system (Figure 10c), while others integrated the two technologies into a single unit (Figure 10d–f). For instance, Song et al. [20] have examined the performance of a full-scale configuration consisting of an MSL system and a CW unit to treat domestic wastewater (Figure 10c). Similarly, Zhang et al. [23] have used the same configuration to treat wastewater produced by agritainment activities (Figure 10c). In terms of design and components (soil, sand, clay, zeolite, and gravel), the two typologies are comparable, but there are some differences, mainly in terms of the size and types of plants used. In addition, both have demonstrated good performance in the removal of pollutants, particularly from domestic wastewater [20], where organic matter and phosphorus removal rates were both about 92%, while nitrogen removal rates were around 76% (Table 1). Furthermore, these eco-friendly combinations have proven to be efficient in mitigating greenhouse gas emissions and required low economic costs.
Koottatep et al. [62], while treating solar septic tank effluent utilizing a pilot plant scale in Thailand (Figure 10d), discovered that this combination was able to remove SS (>70%), organic matter (72–80%), nutrients (>70%), and coliforms (2 log units) (Table 1). In the same context, to treat real rice noodle wastewater in Vietnam, the study conducted by Nguyen et al. [24] using an integrated hybrid system (MSL combined with vertical flow CW) has demonstrated close results for SS (>70%) and organic matter (72–80%) compared to nutrients (54–60%) (Figure 10e). Similarly, in Thailand, Koottatep et al. [25] have investigated the efficacy of a lab-scale system consisting of a vertical flow MSL and CW for treating septic tank wastewater (Figure 10f). The findings showed that this arrangement exceeds 84% in terms of SS and organic matter removal (Table 1). Overall, these three configurations also differ in terms of materials used in the permeable layers (e.g., zeolite, sand, and gravel), the number of soil mixture layers, and the types of plants grown on their surface.
Zhang et al. [51] have previously combined a trickling filter with a horizontal flow MSL to treat domestic wastewater at the lab scale (Figure 10g), and it has demonstrated outstanding efficacy in the elimination of organic matter (≥94%) and nutrients (92–96%) (Table 1). Likewise, Tang et al. [63] have found close results and discovered that this combination (Figure 10g) can process a high content of organic matter (960 mg/L) at varying concentrations of sodium dodecyl benzene sulfonate. Using the same design, Tang et al. [76] have investigated its efficacy in treating mariculture effluent in China. Thus, the authors found that this configuration was also efficient in removing COD and NH4+ (> 80%) (Table 1). In addition, they concluded that extracellular polymer substances could enhance the organic matter removal and mitigate cytotoxicity from salinity.
Luo et al. [26] also designed a two-stage system consisting of a trickling filter and a vertical flow MSL system to treat domestic wastewater in China (Figure 10h). This novel design provides better dispersion of the influent inside the MSL system. Furthermore, the removal efficiencies of COD, TP, NH4+, and TN were 93%, 93%, 86%, and 61%, respectively, at a mean HLR of 920 L/m2/day (Table 1). Although these configurations are comparable in terms of efficiency, they differ in terms of the flow regime, which is vertical in the MSL system operated by Luo et al. [26], compared to the MSL with horizontal flow operated by Zhang et al. [51] and Tang et al. [63,76]. In addition, iron was also used by Luo et al. [26] to support the trickling filter system’s efficiency.
  • Novel hybrid combinations
In the same regard, other studies have examined the efficiency of the MSL combined with other treatment systems. For instance, to treat rural non-point source wastewater, Zhou et al. [9] have developed a novel combination at pilot scale that consists of two stage MSL with blended carbon sources (Figure 10i). This combination also uses ceramsite as a purification material. However, unlike the reviewed combinations, this system is characterized by an upward flow in the two combined units (Figure 10i). In addition, this hybrid system effectively reduced NH4+ (> 50%), TN (64%), and TP (60%) (Table 1). Furthermore, the author concluded that this pilot plant could promote the denitrification process.
Li et al. [21] have developed a novel system consisting of an integrated MSL system and a subsurface wastewater infiltration unit (SWI) that consists of mixed soil and gravel (Figure 10j). Thus, the flow in this combination is in the opposite direction (gravity-flow in the MSL and upward flow in the SWI unit). In addition, this combination recorded a removal rate of 93%, 98%, 74%, and 74% for COD, TP, NH4+, and TN, respectively (Table 1). Latrach et al. [27] have investigated the efficiency of a lab-scale plant consisting of the MSL system followed by a sand filter (Figure 10k). The study findings indicated that the novel conception showed very high efficiency in removing SS (94%), organic matter (80–90%), NH4+ (93%), and TP (78%). In addition, this hybrid combination was effective in removing helminth eggs (100%) at the HLR of 100 L/m2/day (Table 1). Although this system also enhances coliforms and pathogen elimination, the risk of clogging has emerged in this combination.
Overall, the hybrid systems developed, including the MSL system, have been found to be effective in removing pollutants from a variety of wastewater, whether on a lab, pilot, or full scale. This improvement is mostly observed in the reduction of coliform levels (Table 1). For instance, the reduction of coliform content in domestic wastewater has improved from 1.62 log units in single-stage MSL to 3.15 log units in a two-stage MSL system [7]. In addition, it reached 4.80 log units while treating rice noodle wastewater using MSL combined with a CW system [24].
Likewise, the removal of organic matter and phosphorus from domestic wastewater by these reviewed combinations is comparable (Figure 11). However, when it comes to nitrogen removal, this convergence in effectiveness differs, since the two configurations (two-stage MSL and MSL with trickling filter) have high efficiency in eliminating TN when compared to the two systems, MSL-CW and MSL-SWI (Figure 11). In addition, this finding can also be seen in the case of NH4+ removal in the MSL-CW system, suggesting that aeration in these combinations (MSL-CW and MSL-SWI) should be monitored as well as the denitrification process to improve their efficiency in removing nitrogen.

3.6. Multi-Soil-Layering and Alternative Treatment Technologies

This paper also aimed to investigate potential alternative wastewater technologies for MSL through a comprehensive literature analysis. The findings reveal that constructed wetland, lagoon, and sand filter technologies are widely used as rural sewage treatment technologies. Table 2 provides a summary of the comparison’s findings. These alternative methods can be used to treat water in rural regions, according to preliminary comparisons. Additionally, they have the trait of being low-cost and friendly technologies. Thus, the techno-economic evaluation shows that MSL is a suitable option since it can withstand high organic loading rates and only requires a small area for operation. Additionally, MSL can withstand the negative effects of odor or insects. However, MSL’s sanitary efficiency falls short of expectations. Additionally, MSL is an emerging technology and shows a sensitivity to clogging at high HLR. Therefore, conducting more studies in this field has become an urgent necessity to improve MSL’s performance.

3.7. Modeling Approaches Used

Many authors have recently concentrated on simulating the performance of the MSL technology and analyzing linear and non-linear interactions between water quality indicators and MSL efficiencies through modeling approaches. This research axis is considered nowadays as an emerging topic in the MSL field, and researchers have employed a variety of data-driven approaches as well as conventional methods to simulate the performance of the MSL system. Thus, after a thorough analysis of the literature review, regression models and machine learning models were identified as two data-driven models [83] employed to describe MSL behavior, followed by kinetic models. Table 3 presents the number of studies that have targeted modeling water quality indicators in MSL systems, including hybrid systems.

3.7.1. Data-Driven Models

For instance, Sbahi et al. [16] were the first to use neural networks (NN) to predict nitrogen removal in the MSL system (Table 3). Thus, through two hidden layers, the NN model has accurately predicted NH4+, TKN, and TN removal (R2 > 0.93). Furthermore, HLR was proved to be the most significant predictor for all the output variables (p < 0.05). Similarly, for predicting fecal coliform (FC) concentration at a two-stage MSL system’s outlet, other research [6] revealed that the NN model had higher accuracy (R2 ≥ 0.95) compared to the Cubist and multiple linear regression (MLR) methods (R2 = 0.48) (Table 3).
In the same context, the potential of NN in predicting the sanitary efficiency of MSL systems has been previously demonstrated by [11]. Thus, both studies [6,11] revealed that SS and pH were found to be the most important factors in removing coliform bacteria in the MSL system.
For modeling sulfamethoxazole removal, Song et al. [5] compared a stepwise-cluster analysis (SCA) approach with the MLR method. Time and the removal of COD, TP, NH3 (ammonia), NO3, TN, DO (dissolved oxygen), ORP, and pH were all used as input variables in both models. Furthermore, the evaluation of accuracy shows that the MLR model was inaccurate (R2 = 0.62) compared to the SCA model (R2 = 0.82, α < 0.001). In the same context, Hong et al. [13] have used the same approach (SCA) to simulate TN removal in the MSL system (Table 3). The R2 of the best model was greater than 0.93, with a significance level (α = 0.03). Similarly, to investigate the non-linear relationship between pollutant removal and input data in the MSL system, Song et al. [20] have built an SCA model for each output variable (Table 3). The removal rates of COD, BOD5, TP, TN, NH4+, and NO3 were used as input and output data; with time being the only fixed predictor in the proposed input variables. In addition, except for TP (R2 = 0.85) and NO3 (R2 = 0.88), the model precision for the remaining output variables exceeds 0.95 to estimate BOD5 concentrations at the MSL system’s outlet.
On the other hand, data-driven methods were also used for investigating other factors related to wastewater treatment in the MSL system. For instance, Hong et al. [13] have also developed accurate quadratic polynomial functions (Table 3) using PBS, activated sludge, and submerged height as input data, but in this case, to describe the variation of water quality indexes (DO, pH, ORP, removal rates of COD, TP, NH4+, NO3, and TN). In the same vein, through a simple linear model, Nguyen et al. [24] have found that the level of COD and NH4+ at the outlet of the MSL system was expressed with acceptable accuracy (Table 3). However, for the integrated hybrid system (combination of MSL and CW systems), these linear models were more accurate (R2 = 0.92) for both COD removal and COD effluent compared to those for the output variables NH4+ removal and NH4+ effluent (R2 (NH4+effluent) = 0.49, R2 (NH4+removal) = 0.58), respectively.

3.7.2. Kinetic Models

Regarding kinetic models, Tang et al. [76] have proposed two logistic kinetic functions to calculate the removal of COD and NH4+ in the combination that consisted of a vertical flow trickling filter associated with a horizontal flow MSL system. For the final effluent of COD, the model input variable was only salinity as an external environmental factor, while the concentration of NH4+ in the effluent was related to BOD5 load, media surface area, and filter effluent temperature, respectively. Both models had a high level of accuracy (R2 = 0.94). Koottatep et al. [25], in turn, investigated both the completely mixed flow and plug flow models based on COD, BOD5, and NH3 concentrations (Table 3). Thus, the suitable kinetic model and coefficients were based on the highest accuracy. The findings indicate that the first-order completely mixed model was proposed as the best fit model for the MSL-vertical flow CW system for the three output variables (0.98 ≤ R2 ≤ 0.98) as indicated in Table 3.
In this study, the methods with regression structures are allocated to the regression model category, whereas approaches involving undefined model structures are given to the machine learning model category [83]. Figure 12 also shows the distribution of the three common modeling approaches among the various modeled water quality indicators in the MSL systems. Thus, nitrogen was the most modeled parameter in seven studies with 18 models (nine using regression, seven using machine learning, and two using kinetic models), followed by COD with seven models using the three approaches, while phosphorus modeling using regression and machine learning has been the subject of two studies (Figure 12). This significant attention might be due to the large amount of research effort spent on nitrogen removal in decentralized wastewater treatment systems [41]. Models for coliform bacteria have also been developed recently, as illustrated in Figure 12, although with less applicability than the other parameters discussed above.
In summary, NN and SCA dominate machine learning models in MSL performance modeling. Thus, the models have been used to predict MSL performance, investigate its behavior, and analyze non-linear relationships between water quality indicators and operational variables such as time and HLR. Furthermore, these models have demonstrated moderate to high prediction accuracy when it comes to simulating real MSL performance. Although this literature review considers machine learning models to be credible approaches for simulating MSL system performance, their application in the MSL field is still relatively new, with only 11 research papers among 70. However, the results are still encouraging and might be useful to the MSL system in the future. Similarly, while kinetic models are crucial in the research of treatment systems as well as the management and control of their efficacy, they are not up to the task. Future research in this area is also needed to better understand MSL behavior and increase its efficacy.

4. Recommendations and Future Consideration

For three decades, various research studies have been carried out in the field of multi-soil-Layering (MSL) technology, but still further research is required as per the review’s outcome. Although the MSL system has shown its effectiveness as a nature-based technology to treat wastewater under different operating conditions, its use is still very limited on a large scale. Further research is still needed to enhance the treatment efficiency of MSL and optimize its design criteria. Therefore, through this bibliometric analysis and review, we will try to evaluate this technology in comparison with similar treatment technologies, point out certain knowledge gaps and challenges, and suggest some recommendations to help researchers gain a better understanding of MSL behavior.

4.1. MSL Treatment Efficiency

In this literature review, the majority of studies focused on investigating contaminant removal in the MSL system. However, there is some gap in knowledge about MSL behavior and its sanitation efficiency. Thus, there are urgent challenges in improving MSL’s performance against pathogens and viral indicators in wastewater. Thus, this review urges investigators to investigate predatory activities and microbial flora within the MSL system biofilm responsible for removal and transformation mechanisms. In addition, the findings of this review also encouraged researchers to address other topics that are not covered, such as highlighting factors that influence nitrification and denitrification processes as well as phosphorus sorption dynamics; evaluating the reuse of MSL effluents on irrigation and soil properties; studying the kinetics of contaminant removal in the MSL technology; investigating the effect of MSL characteristic media and residence time on microorganism removal; and evaluating the MSL effectiveness under variation of the hydraulic and organic regime, aeration conditions and type of substrate. In the same context, further research regarding the long-term pollution removal performance of this system is needed. Future research by MSLs should also consider the mechanisms involved in the removal of certain pollutants, such as emerging pollutants, heavy metals, and toxins.

4.2. Design and Costing

The authors have extensive expertise in MSL technology and have several papers in peer-reviewed scientific journals. However, the MSL system is still considered an emerging solution for wastewater treatment, which calls for continuous improvements. This paper recommends developing a new composition of soil mixture to improve MSL purification performance by adding mineral apatite and smectite clays as high purifying materials [101,102]. Furthermore, the majority of papers investigated in this review tested MSL systems with vertical flow. Thus, additional studies on MSL purification efficiencies under horizontal flow and with a hybrid system are recommended. In addition, there is a lack of sufficient knowledge about the effect of used substrates and their structures within the MSL on the removal efficiency.
On the other hand, although a recent study [68] attempted to estimate the cost of soil mixture blocks in the lab and in the field, which is between USD 0.35 and USD 3 per packet, as well as the cost of zeolite (about USD 20 per kg). However, more research into this aspect of the MSL system is needed, especially because material prices are still fluctuating all around the world, and per capita income is low, particularly in rural parts of developing countries. In addition, estimating the MSL’s initial investment (including sizing), water quality enhancement, operation, and maintenance are required for a full-scale MSL.

4.3. MSL Modeling

Regarding modeling approaches to simulate the performance of the MSL system; most of the models developed recently are constrained and cannot describe the non-linear interactions between the water indicators within the MSL system. Thus, the use of more complex models (e.g., random forest, support vector machine) is recommended, especially with the advent of artificial intelligence techniques. Moreover, the majority of the models have mainly investigated input/output datasets rather than MSL internal processes datasets. Nevertheless, kinetic models are also important to understand pollutant removal in this system. In addition, determining the MSL’s kinetic constant may aid in the optimization of the MSL system’s investment cost. Regarding numerical solutions, models that describe the growth and death of bacterial species (e.g., autotrophs and heterotrophs) might potentially aid in understanding MSL behavior in terms of the removal mechanisms. On the other hand, it is suggested that the authors use mechanistic models to investigate the MSL system’s physical (dispersion-advection), chemical, and biological processes.
Another issue to be developed in the future is the environmental impact assessment of different applications involving MSLs in wastewater treatment to find solutions to improve their environmental performances. Thus, elaborating guidelines providing useful information on how to implement, operate, and maintain MSL nature-based technology is actually needed.

5. Conclusions

The multi-soil-layering (MSL) system is considered an emerging ecotechnology that has proven its efficiency in the treatment of a wide range of wastewater across the world (97% for SS, 98% for BOD5, and 100% for NH4+ and TP). This technology is characterized by several advantages, such as low investment costs, small area occupation, simple maintenance, and no energy requirements, which make it an interesting option to be integrated with other technologies to improve its performance. However, among 70 publications, only 14 research papers were found in this literature review combining MSL with other wastewater treatment methods (e.g., constructed wetlands, trickling filter, sand filter, etc.). Although numerous scientific studies have looked into and monitored the MSL system, the majority of these studies have been done on a lab or pilot scale. Thus, full-scale MSL systems should be an area for future research to evaluate their behavior in a real environment rather than on a lab scale. Thus, this will allow for a better understanding of the challenges the system may face and an increase in its sanitary efficiency particularly (1.62 log units in single-stage MSL). Other aspects of the research that must be investigated include the reuse of the MSL effluents in the agricultural field and the irrigation of green space, as many countries that have implemented this technology suffer from water scarcity, especially with the lack of precipitation and consideration of climate change. Furthermore, it is important to investigate the kinetic of the MSL systems and further explore non-linear relationships between water quality parameters in the MSL, mainly fecal indicators bacteria and pathogens, to better understand their removal process. On the other hand, the addition of a carbon source within the soil mixture can hasten the growth and diversity of the microbial community, hence improving the MSL system’s efficiency. Published studies on the potential financial cost of the MSL system are still limited, and they have not allowed for the development of first sizing rules for vertical and/or horizontal flow MSL systems used as secondary or tertiary treatment (admissible pollutant load, surface occupation, investment cost/inhabitant). Therefore, attention to the challenges and recommendations proposed in this review will inevitably contribute to the improvement of MSL sanitation efficiency, cost estimation, and simulation of its behavior with high accuracy (R2 > 90%).

Author Contributions

Investigation, methodology, visualization, software, writing original draft, S.S.; Review, editing and improved the writing, N.O., T.M., and A.H.; Conceptualization, investigation, supervision, Review, editing and improved the writing, L.M. All authors have read and agreed to the published version of the manuscript.


This research received no funding.

Data Availability Statement

The data presented in this paper are available in the Web of Science database (, accessed on 1 June 2022).

Conflicts of Interest

The authors declare no conflict of interest.


BCSblended organic carbon source
BOD5five-day biochemical oxygen demand
CO2carbon dioxide
CODchemical oxygen demand
CWconstructed wetland
DOdissolved oxygen
ECelectrical conductivity
FCfecal coliform
Fe0metallic iron
Fe2+ferrous ions
Fe3+ferric ions
FIBfecal bacteria indicator
HLRhydraulic loading rate
MLRmultiple linear regression
N2Onitrous oxide
NNneural networks
ORPoxidation-reduction potential
PBSpolybutylene succinate
PLpermeable layer
SCAstepwise-cluster analysis
SFCWsubsurface flow constructed wetland
SSsuspended solids
SWIsubsurface wastewater infiltration
TCtotal coliform
TNtotal nitrogen
TKNtotal kjeldahl nitrogen
TPtotal phosphorus
VFCWvertical flow constructed wetland
WoSWeb of Science


  1. Sato, K.; Wakatsuki, T.; Iwashima, N.; Masunaga, T. Evaluation of Long-Term Wastewater Treatment Performances in Multi-Soil-Layering Systems in Small Rural Communities. Appl. Environ. Soil Sci. 2019, 2019, 11. [Google Scholar] [CrossRef]
  2. Moreira, F.D.; Dias, E.H.O. Constructed Wetlands Applied in Rural Sanitation: A Review. Environ. Res. 2020, 190, 110016. [Google Scholar] [CrossRef] [PubMed]
  3. WHO. Guidelines on Sanitation and Health; World Health Organization: Geneva, Switzerland, 2018; pp. 1–220. Available online: (accessed on 20 June 2022).
  4. Yaqoob, A.A.; Parveen, T.; Umar, K.; Mohamad Ibrahim, M.N. Role of Nanomaterials in the Treatment of Wastewater: A Review. Water 2020, 12, 495. [Google Scholar] [CrossRef]
  5. Song, P.; Huang, G.; An, C.; Xin, X.; Zhang, P.; Chen, X.; Ren, S.; Xu, Z.; Yang, X. Exploring the Decentralized Treatment of Sulfamethoxazole-Contained Poultry Wastewater through Vertical-Flow Multi-Soil-Layering Systems in Rural Communities. Water Res. 2021, 188, 116480. [Google Scholar] [CrossRef]
  6. Sbahi, S.; Ouazzani, N.; Hejjaj, A.; Mandi, L. Neural Network and Cubist Algorithms to Predict Fecal Coliform Content in Treated Wastewater by Multi-Soil-Layering System for Potential Reuse. J. Environ. Qual. 2021, 50, 144–157. [Google Scholar] [CrossRef]
  7. Latrach, L.; Ouazzani, N.; Hejjaj, A.; Mahi, M.; Masunaga, T.; Mandi, L. Two-Stage Vertical Flow Multi-Soil-Layering (MSL) Technology for Efficient Removal of Coliforms and Human Pathogens from Domestic Wastewater in Rural Areas under Arid Climate. Int. J. Hyg. Environ. Health 2018, 221, 64–80. [Google Scholar] [CrossRef]
  8. Idris, M.O.; Yaqoob, A.A.; Ibrahim, M.N.M.; Ahmad, A.; Alshammari, M.B. Introduction of Adsorption Techniques for Heavy Metals Remediation. In Emerging Techniques for Treatment of Toxic Metals from Wastewater, 1st ed.; Elsevier: Amsterdam, The Netherlands, 2023; pp. 1–18. [Google Scholar]
  9. Zhou, Q.; Sun, H.; Jia, L.; Zhao, L.; Wu, W. Enhanced Pollutant Removal from Rural Non-Point Source Wastewater Using a Two-Stage Multi-Soil-Layering System with Blended Carbon Sources: Insights into Functional Genes, Microbial Community Structure and Metabolic Function. Chemosphere 2021, 275, 130007. [Google Scholar] [CrossRef]
  10. Song, P.; Huang, G.; An, C.; Shen, J.; Zhang, P.; Chen, X.; Shen, J.; Yao, Y.; Zheng, R.; Sun, C. Treatment of Rural Domestic Wastewater Using Multi-Soil-Layering Systems: Performance Evaluation, Factorial Analysis and Numerical Modeling. Sci. Total Environ. 2018, 644, 536–546. [Google Scholar] [CrossRef]
  11. Sbahi, S.; Ouazzani, N.; Latrach, L.; Hejjaj, A.; Mandi, L. Predicting the Concentration of Total Coliforms in Treated Rural Domestic Wastewater by Multi-Soil-Layering (MSL) Technology Using Artificial Neural Networks. Ecotoxicol. Environ. Saf. 2020, 204, 111118. [Google Scholar] [CrossRef]
  12. An, C.J.; McBean, E.; Huang, G.H.; Yao, Y.; Zhang, P.; Chen, X.J.; Li, Y.P. Multi-Soil-Layering Systems for Wastewater Treatment in Small and Remote Communities. J. Environ. Inform. 2016, 27, 131–144. [Google Scholar] [CrossRef]
  13. Hong, Y.; Huang, G.; An, C.; Song, P.; Xin, X.; Chen, X.; Zhang, P.; Zhao, Y.; Zheng, R. Enhanced Nitrogen Removal in the Treatment of Rural Domestic Sewage Using Vertical-Flow Multi-Soil-Layering Systems: Experimental and Modeling Insights. J. Environ. Manag. 2019, 240, 273–284. [Google Scholar] [CrossRef] [PubMed]
  14. Latrach, L.; Masunaga, T.; Ouazzani, N.; Hejjaj, A.; Mahi, M.; Mandi, L. Removal of Bacterial Indicators and Pathogens from Domestic Wastewater by the Multi-Soil-Layering (MSL) System. Soil Sci. Plant Nutr. 2015, 61, 337–346. [Google Scholar] [CrossRef]
  15. Aba, R.P.; Mugani, R.; Hejjaj, A.; Brugerolle de Fraissinette, N.; Oudra, B.; Ouazzani, N.; Campos, A.; Vasconcelos, V.; Carvalho, P.N.; Mandi, L. First Report on Cyanotoxin (MC-LR) Removal from Surface Water by Multi-Soil-Layering (MSL) Eco-Technology: Preliminary Results. Water 2021, 13, 1403. [Google Scholar] [CrossRef]
  16. Sbahi, S.; Ouazzani, N.; Hejjaj, A.; Mandi, L. Nitrogen Modeling and Performance of Multi-Soil-Layering (MSL) Bioreactor Treating Domestic Wastewater in Rural Community. J. Water Process Eng. 2021, 44, 102389. [Google Scholar] [CrossRef]
  17. Shen, J.; Huang, G.; An, C.; Song, P.; Xin, X.; Yao, Y.; Zheng, R. Biophysiological and Factorial Analyses in the Treatment of Rural Domestic Wastewater Using Multi-Soil-Layering Systems. J. Environ. Manag. 2018, 226, 83–94. [Google Scholar] [CrossRef] [PubMed]
  18. Zidan, K.; Sbahi, S.; Hejjaj, A.; Ouazzani, N.; Assabbane, A.; Mandi, L. Removal of Bacterial Indicators in On-Site Two-Stage Multi-Soil-Layering Plant under Arid Climate (Morocco): Prediction of Total Coliform Content Using K-Nearest Neighbor Algorithm. Environ. Sci. Pollut. Res. 2022, 29, 75716–75729. [Google Scholar] [CrossRef]
  19. Sato, K.; Iwashima, N.; Wakatsuki, T.; Masunaga, T. Quantitative Evaluation of Treatment Processes and Mechanisms of Organic Matter, Phosphorus, and Nitrogen Removal in a Multi-Soil-Layering System. Soil Sci. Plant Nutr. 2011, 57, 475–486. [Google Scholar] [CrossRef]
  20. Song, P.; Huang, G.; An, C.; Zhang, P.; Chen, X.; Ren, S. Performance Analysis and Life Cycle Greenhouse Gas Emission Assessment of an Integrated Gravitational-Flow Wastewater Treatment System for Rural Areas. Environ. Sci. Pollut. Res. 2019, 26, 25883–25897. [Google Scholar] [CrossRef]
  21. Li, D.; Wang, X.; Chi, L.; Zhang, Z.; Liu, Y.; Li, X. Decentralized Domestic Sewage Treatment Using an Integrated Multi-Soil-Layering and Subsurface Wastewater Infiltration System. Water 2021, 13, 431. [Google Scholar] [CrossRef]
  22. Maeng, S.K.; Park, J.W.; Noh, J.H.; Won, S.Y.; Song, K.G. Dissolved Organic Matter Characteristics and Removal of Trace Organic Contaminants in a Multi-Soil-Layering System. J. Environ. Chem. Eng. 2021, 9, 105446. [Google Scholar] [CrossRef]
  23. Zhang, P.; Huang, G.; An, C.; Fu, H.; Gao, P.; Yao, Y.; Chen, X. An Integrated Gravity-Driven Ecological Bed for Wastewater Treatment in Subtropical Regions: Process Design, Performance Analysis, and Greenhouse Gas Emissions Assessment. J. Clean. Prod. 2019, 212, 1143–1153. [Google Scholar] [CrossRef]
  24. Nguyen, X.C.; Chang, S.W.; Tran, T.C.P.; Nguyen, T.T.N.; Hoang, T.Q.; Banu, J.R.; Al-Muhtaseb, A.H.; La, D.D.; Guo, W.; Ngo, H.H. Comparative Study about the Performance of Three Types of Modified Natural Treatment Systems for Rice Noodle Wastewater. Bioresour. Technol. 2019, 282, 163–170. [Google Scholar] [CrossRef] [PubMed]
  25. Koottatep, T.; Suksiri, P.; Pussayanavin, T.; Polprasert, C. Development of a Novel Multi-Soil Layer Constructed Wetland Treating Septic Tank Effluent with Emphasis on Organic and Ammonia Removals. Water Air Soil Pollut. 2018, 229, 258. [Google Scholar] [CrossRef]
  26. Luo, W.; Yang, C.; He, H.; Zeng, G.; Yan, S.; Cheng, Y. Novel Two-Stage Vertical Flow Biofilter System for Efficient Treatment of Decentralized Domestic Wastewater. Ecol. Eng. 2014, 64, 415–423. [Google Scholar] [CrossRef]
  27. Latrach, L.; Ouazzani, N.; Masunaga, T.; Hejjaj, A.; Bouhoum, K.; Mahi, M.; Mandi, L. Domestic Wastewater Disinfection by Combined Treatment Using Multi-Soil-Layering System and Sand Filters (MSL-SF): A Laboratory Pilot Study. Ecol. Eng. 2016, 91, 294–301. [Google Scholar] [CrossRef]
  28. Wu, J.; Wang, Z. A Hybrid Model for Water Quality Prediction Based on an Artificial Neural Network, Wavelet Transform, and Long Short-Term Memory. Water 2022, 14, 610. [Google Scholar] [CrossRef]
  29. Alnahit, A.O.; Mishra, A.K.; Khan, A.A. Stream Water Quality Prediction Using Boosted Regression Tree and Random Forest Models. Stoch. Environ. Res. Risk Assess. 2022, 36, 2661–2680. [Google Scholar] [CrossRef]
  30. Ho, C.C.; Wang, P.H. Efficiency of a Multi-Soil-Layering System on Wastewater Treatment Using Environment-Friendly Filter Materials. Int. J. Environ. Res. Public Health 2015, 12, 3362–3380. [Google Scholar] [CrossRef]
  31. Luanmanee, S.; Attanandana, T.; Masunaga, T.; Wakatsuki, T. The Efficiency of a Multi-Soil-Layering System on Domestic Wastewater Treatment during the Ninth and Tenth Years of Operation. Ecol. Eng. 2001, 18, 185–199. [Google Scholar] [CrossRef]
  32. Masunaga, T.; Sato, K.; Mori, J.; Shirahama, M.; Kudo, H.; Wakatsuki, T. Characteristics of Wastewater Treatment Using a Multi-Soil-Layering System in Relation to Wastewater Contamination Levels and Hydraulic Loading Rates. Soil Sci. Plant Nutr. 2007, 53, 215–223. [Google Scholar] [CrossRef]
  33. Khalifa, J.; Ouazzani, N.; Hejjaj, A.; Mandi, L. Remediation and Disinfection Capabilities Assessment of Some Local Materials to Be Applied in Multi-Soil-Layering (MSL) Ecotechnology. Desalin. Water Treat. 2020, 178, 53–64. [Google Scholar] [CrossRef]
  34. Liu, C.; Huang, G.; Song, P.; An, C.; Zhang, P.; Shen, J.; Ren, S.; Zhao, K.; Huang, W.; Xu, Y. Treatment of Decentralized Low-Strength Livestock Wastewater Using Microcurrent-Assisted Multi-Soil-Layering Systems: Performance Assessment and Microbial Analysis. Chemosphere 2022, 294, 133536. [Google Scholar] [CrossRef] [PubMed]
  35. Luanmanee, S.; Boonsook, P.; Attanandana, T.; Wakatsuki, T. Effect of Organic Components and Aeration Regimes on the Efficiency of a Multi-Soil-Layering System for Domestic Wastewater Treatment. Soil Sci. Plant Nutr. 2002, 48, 125–134. [Google Scholar] [CrossRef]
  36. Guan, Y.; Zhang, Y.; Zhong, C.N.; Huang, X.F.; Fu, J.; Zhao, D. Effect of Operating Factors on the Contaminants Removal of a Soil Filter: Multi-Soil-Layering System. Environ. Earth Sci. 2015, 74, 2679–2686. [Google Scholar] [CrossRef]
  37. Latrach, L.; Ouazzani, N.; Hejjaj, A.; Zouhir, F.; Mahi, M.; Masunaga, T.; Mandi, L. Optimization of Hydraulic Efficiency and Wastewater Treatment Performances Using a New Design of Vertical Flow Multi-Soil-Layering (MSL) Technology. Ecol. Eng. 2018, 117, 140–152. [Google Scholar] [CrossRef]
  38. Zhao, L.; Dai, T.; Qiao, Z.; Sun, P.; Hao, J.; Yang, Y. Application of Artificial Intelligence to Wastewater Treatment: A Bibliometric Analysis and Systematic Review of Technology, Economy, Management, and Wastewater Reuse. Process Saf. Environ. Prot. 2020, 133, 169–182. [Google Scholar] [CrossRef]
  39. Dalpé, R. Bibliometric Analysis of Biotechnology. Scientometrics 2002, 55, 189–213. [Google Scholar] [CrossRef]
  40. Xie, H.; Zhang, Y.; Wu, Z.; Lv, T. A Bibliometric Analysis on Land Degradation: Current Status, Development, and Future Directions. Land 2020, 9, 28. [Google Scholar] [CrossRef]
  41. Wu, L.; Wang, W.; Jing, P.; Chen, Y.; Zhan, F.; Shi, Y.; Li, T. Travel Mode Choice and Their Impacts on Environment—A Literature Review Based on Bibliometric and Content Analysis, 2000–2018. J. Clean. Prod. 2020, 249, 119391. [Google Scholar] [CrossRef]
  42. Van Eck, N.J.; Waltman, L. Software Survey: VOSviewer, a Computer Program for Bibliometric Mapping. Scientometrics 2010, 84, 523–538. [Google Scholar] [CrossRef]
  43. Chen, X.; Sato, K.; Wakatsuki, T.; Masunaga, T. Effect of Structural Difference on Wastewater Treatment Efficiency in Multi-Soil-Layering Systems: Relationship between Soil Mixture Block Size and Removal Efficiency of Selected Contaminants. Soil Sci. Plant Nutr. 2007, 53, 206–214. [Google Scholar] [CrossRef]
  44. Paul, E. Soil Microbiology, Ecology and Biochemistry, 4th ed.; Academic Press: San Diego, CA, USA, 2006; p. 552. [Google Scholar]
  45. Masunaga, T.; Sato, K.; Senga, Y.; Seike, Y.; Inaishi, T.; Kudo, H.; Wakatsuki, T. Characteristics of CO2, CH4 and N2O Emissions from a Multi-Soil-Layering System during Wastewater Treatment. Soil Sci. Plant Nutr. 2007, 53, 173–180. [Google Scholar] [CrossRef]
  46. Stevik, T.K.; Aa, K.; Ausland, G.; Hanssen, J.F. Retention and Removal of Pathogenic Bacteria in Wastewater Percolating through Porous Media: A Review. Water Res. 2004, 38, 1355–1367. [Google Scholar] [CrossRef] [PubMed]
  47. Kadam, A.; Oza, G.; Nemade, P.; Dutta, S.; Shankar, H. Municipal Wastewater Treatment Using Novel Constructed Soil Filter System. Chemosphere 2008, 71, 975–981. [Google Scholar] [CrossRef] [PubMed]
  48. Song, P.; Huang, G.; Hong, Y.; An, C.; Xin, X.; Zhang, P. A Biophysiological Perspective on Enhanced Nitrate Removal from Decentralized Domestic Sewage Using Gravitational-Flow Multi-Soil-Layering Systems. Chemosphere 2020, 240, 124868. [Google Scholar] [CrossRef] [PubMed]
  49. Wei, C.; Wu, W. Performance of Single-Pass and by-Pass Multi-Step Multi-Soil-Layering Systems for Low-(C/N)-Ratio Polluted River Water Treatment. Chemosphere 2018, 206, 579–586. [Google Scholar] [CrossRef]
  50. Song, Y.; Huang, Y.T.; Ji, H.F.; Nie, X.J.; Zhang, Z.Y.; Ge, C.; Luo, A.C.; Chen, X. Treatment of Turtle Aquaculture Effluent by an Improved Multi-Soil-Layer System. J. Zhejiang Univ. Sci. B 2015, 16, 145–154. [Google Scholar] [CrossRef]
  51. Zhang, Y.; Cheng, Y.; Yang, C.; Luo, W.; Zeng, G.; Lu, L. Performance of System Consisting of Vertical Flow Trickling Filter and Horizontal Flow Multi-Soil-Layering Reactor for Treatment of Rural Wastewater. Bioresour. Technol. 2015, 193, 424–432. [Google Scholar] [CrossRef]
  52. Sato, K.; Masunaga, T.; Wakatsuki, T. Water Movement Characteristics in a Multi-Soil-Layering System. Soil Sci. Plant Nutr. 2005, 51, 75–82. [Google Scholar] [CrossRef]
  53. Zein, R.; Ningsih, Z.S.; Novita, L.; Swesty, N.; Mukhlis; Novrian, H. Treatment of Waste Water Noodle Industry with a Multi-Soil-Layering (MSL) System. Res. J. Pharm. Biol. Chem. Sci. 2016, 7, 88–94. [Google Scholar]
  54. Yidong, G.; Xin, C.; Shuai, Z.; Ancheng, L. Performance of Multi-Soil-Layering System (MSL) Treating Leachate from Rural Unsanitary Landfills. Sci. Total Environ. 2012, 420, 183–190. [Google Scholar] [CrossRef] [PubMed]
  55. Sato, K.; Masunaga, T.; Wakatsuki, T. Characterization of Treatment Processes and Mechanisms of COD, Phosphorus and Nitrogen Removal in a Multi-Soil-Layering System. Soil Sci. Plant Nutr. 2005, 51, 213–221. [Google Scholar] [CrossRef]
  56. Chen, X.; Luo, A.C.; Sato, K.; Wakatsuki, T.; Masunaga, T. An Introduction of a Multi-Soil-Layering System: A Novel Green Technology for Wastewater Treatment in Rural Areas. Water Environ. J. 2009, 23, 255–262. [Google Scholar] [CrossRef]
  57. Guo, J.; Zhou, Y.; Yang, Y.; Chen, C.; Xu, J. Effects of Hydraulic Loading Rate on Nutrients Removal from Anaerobically Digested Swine Wastewater by Multi Soil Layering Treatment Bioreactor. Int. J. Environ. Res. Public Health 2018, 15, 2688. [Google Scholar] [CrossRef]
  58. Sato, K.; Iwashima, N.; Wakatsuki, T.; Masunaga, T. Clarification of Water Movement Properties in a Multi-Soil-Layering System. Soil Sci. Plant Nutr. 2011, 57, 607–618. [Google Scholar] [CrossRef]
  59. Guo, J.; Zhou, Y.; Jiang, S.; Chen, C. Feasibility Investigation of a Multi Soil Layering Bioreactor for Domestic Wastewater Treatment. Environ. Technol. 2019, 40, 2317–2324. [Google Scholar] [CrossRef]
  60. Masunaga, T.; Sato, K.; Zennami, T.; Fujii, S.; Wakatsuki, T. Direct Treatment of Polluted River Water by the Multi-Soil-Layering Method. J. Water Environ. Technol. 2003, 1, 97–104. [Google Scholar] [CrossRef]
  61. Guan, Y.D.; Xu, D.F.; Chen, X.; Luo, A.C.; Fang, H.; Song, Y.Z. Flow Patterns of Multi-Soil-Layering Systems. Desalination Water Treat. 2014, 52, 4165–4169. [Google Scholar] [CrossRef]
  62. Koottatep, T.; Pussayanavin, T.; Khamyai, S.; Polprasert, C. Performance of Novel Constructed Wetlands for Treating Solar Septic Tank Effluent. Sci. Total Environ. 2021, 754, 142447. [Google Scholar] [CrossRef]
  63. Tang, W.; Li, X.; Liu, H.; Wu, S.; Zhou, Q.; Du, C.; Teng, Q.; Zhong, Y.; Yang, C. Sequential Vertical Flow Trickling Filter and Horizontal Flow Multi-Soil-Layering Reactor for Treatment of Decentralized Domestic Wastewater with Sodium Dodecyl Benzene Sulfonate. Bioresour. Technol. 2020, 300, 122634. [Google Scholar] [CrossRef]
  64. Ait-Hmane, A.; Mandi, L.; Ouazzani, N.; Ait Hammou, H.; Hejjaj, A.; Alahiane, S.; Assabbane, A. Combined Treatment of Olive Mill Wastewater by Multi-Soil-Layering Ecotechnology and Adsorption on Activated Carbon/Lime. Desalination Water Treat. 2021, 233, 253–260. [Google Scholar] [CrossRef]
  65. Wang, J.; Zhang, Z.; Guo, Y.; Zhang, L.; Liu, J. Experimental Study on the Treatment of Rural Domestic Wastewater Using the Multi-Soil-Layering System Filled with Sludge-Based Biochar. Ann. Chim. Sci. Mater. 2021, 45, 161–165. [Google Scholar] [CrossRef]
  66. Sy, S.; Sofyan, S.; Ardinal, A.; Kasman, M. Reduction of Pollutant Parameters in Textile Dyeing Wastewater by Gambier (Uncaria gambir Roxb) Using the Multi Soil Layering (MSL) Bioreactor. Mater. Sci. Eng. 2019, 546, 22–32. [Google Scholar] [CrossRef]
  67. Xiao, M.; Hu, R.; Cui, X.; Gwenzi, W.; Noubactep, C. Understanding the Operating Mode of Fe0/Fe-Sulfide/H2O Systems for Water Treatment. Processes 2020, 8, 409. [Google Scholar] [CrossRef]
  68. Chen, Y.C.; Pat, H.W. Comparing Natural Red Soil and Irons for Removal of Phosphorus from Wastewater Using the Multi-Soil-Layering System and Its Economic Analysis. J. Environ. Manag. 2021, 296, 113252. [Google Scholar] [CrossRef]
  69. Liu, D.; Li, J.; Li, C.; Deng, Y.; Zhang, Z.; Ye, Z.; Zhu, S. Poly(Butylene Succinate)/Bamboo Powder Blends as Solid-Phase Carbon Source and Biofilm Carrier for Denitrifying Biofilters Treating Wastewater from Recirculating Aquaculture System. Sci. Rep. 2018, 8, 3289. [Google Scholar] [CrossRef]
  70. Pattnaik, R.; Yost, R.S.; Porter, G.; Masunaga, T.; Attanandana, T. Improving Multi-Soil-Layer (MSL) System Remediation of Dairy Effluent. Ecol. Eng. 2008, 32, 1–10. [Google Scholar] [CrossRef]
  71. Colella, C. Natural zeolites in environmentally friendly processes and applications. Stud. Surf. Sci. Catal. 1999, 125, 641–655. [Google Scholar]
  72. Boonsook, P.; Luanmanee, S.; Attanandana, T.; Kamidouzono, A.; Masunaga, T.; Wakatsuki, T.; Luanmanee, S.; Kamidouzono, A.; Masunaga, T.; Wakatsuki, T. A Comparative Study of Permeable Layer Materials and Aeration Regime on Efficiency of Multi-Soil-Layering System for Domestic Wastewater Treatment in Thailand. Soil Sci. Plant Nutr. 2003, 49, 873–882. [Google Scholar] [CrossRef]
  73. Chen, X.; Sato, K.; Wakatsuki, T.; Masunaga, T. Effect of Aeration and Material Composition in Soil Mixture Block on the Removal of Colored Substances and Chemical Oxygen Demand in Livestock Wastewater Using Multi-Soil-Layering Systems. Soil Sci. Plant Nutr. 2007, 53, 509–516. [Google Scholar] [CrossRef]
  74. Attanandana, T.; Saitthiti, B.; Thongpae, S.; Kritapirom, S.; Luanmanee, S.; Wakatsuki, T. Multi-Media-Layering System for Food Service Wastewater Treatment. Ecol. Eng. 2000, 15, 133–138. [Google Scholar] [CrossRef]
  75. Luanmanee, S.; Boonsook, P.; Attanandana, T.; Saitthiti, B.; Panichajakul, C.; Wakatsuki, T. Effect of Intermittent Aeration Regulation of a Multi-Soil-Layering System on Domestic Wastewater Treatment in Thailand. Ecol. Eng. 2002, 18, 415–428. [Google Scholar] [CrossRef]
  76. Tang, W.; Wu, M.; Lou, W.; Yang, C. Role of Extracellular Polymeric Substances and Enhanced Performance for Biological Removal of Carbonaceous Organic Matters and Ammonia from Wastewater with High Salinity and Low Nutrient Concentrations. Bioresour. Technol. 2021, 326, 124764. [Google Scholar] [CrossRef] [PubMed]
  77. Abourida, A. Approche Hydrogéologique de la Nappe du Haouz (Maroc) par Télédétection, Isotopie, SIG et Modélisation. Ph.D. Thesis, Université Cadi Ayyad Faculté des Sciences et Techniques, Marrakech, Morocco, 2007; p. 146. [Google Scholar]
  78. Zema, D.A.; Calabro, P.S.; Folino, A.; Tamburino, V.; Zappia, G.; Zimbone, S.M. Wastewater Management in Citrus Processing Industries: An Overview of Advantages and Limits. Water 2019, 11, 2481. [Google Scholar] [CrossRef]
  79. Almasia, A.; Mohammadib, M.; Salehniac, S.; Azizib, M.H.; Pirsahebb, M.; Zolfagharid, M.R. Removal of Parasitic Particles, Protozoa Cysts, and Thermotolerant Coliforms in the Integrated Aeration Lagoon, Case Study: Iran. Desalin. Water Treat. 2019, 137, 221–225. [Google Scholar] [CrossRef]
  80. Kauppinen, A.; Martikainen, K.; Matikka, V.; Veijalainen, A.-M.; Pitkänen, T.; Heinonen-Tanski, H.; Miettinen, I.T. Sand Filters for Removal of Microbes and Nutrients from Wastewater during a One-Year Pilot Study in a Cold Temperate Climate. J. Environ. Manag. 2014, 133, 206–213. [Google Scholar] [CrossRef]
  81. Chen, S.; Chen, Z.; Dougherty, M.; Zuo, X.; He, J. The Role of Clogging in Intermittent Sand Filter (ISF) Performance in Treating Rural Wastewater Retention Pond Effluent. J. Clean. Prod. 2021, 294, 126309. [Google Scholar] [CrossRef]
  82. Guo, L.K.; Yang, L.; Ren, Y.X.; Dou, J.W.; Cui, S.; Lan, J.; Li, X.T.; Wang, J.; Wang, Y.C. Enhanced Biofilm Formation and Denitrification in Slow Sand Filters for Advanced Nitrogen Removal by Powdery Solid Carbon Sources Addition. J. Water Process Eng. 2022, 50, 103192. [Google Scholar] [CrossRef]
  83. Jia, Y.; Zheng, F.; Maier, H.R.; Ostfeld, A.; Creaco, E.; Savic, D.; Langeveld, J.; Kapelan, Z. Water Quality Modelling in Sewer Networks: Review and Future Research Directions. Water Res. 2021, 202, 117419. [Google Scholar] [CrossRef]
  84. Kuhn, M.; Johnson, K. Applied Predictive Modeling, 1st ed.; Springer: New York, NY, USA, 2013; p. 615. [Google Scholar]
  85. Wang, X.; Huang, G.; Lin, Q.; Nie, X.; Cheng, G.; Fan, Y.; Li, Z.; Yao, Y.; Suo, M. A Stepwise Cluster Analysis Approach for Downscaled Climate Projection–A Canadian Case Study. Environ. Model. Softw. 2013, 49, 141–151. [Google Scholar] [CrossRef]
  86. Zhuang, X.W.; Li, Y.P.; Huang, G.H.; Liu, J. Assessment of Climate Change Impacts on Watershed in Cold-Arid Region: An Integrated Multi-GCM-Based Stochastic Weather Generator and Stepwise Cluster Analysis Method. Clim. Dyn. 2016, 47, 191–209. [Google Scholar] [CrossRef]
  87. Elmaz, F.; Yücel, Ö.; Mutlu, A.Y. Predictive Modeling of Biomass Gasification with Machine Learning-Based Regression Methods. Energy 2020, 191, 116541. [Google Scholar] [CrossRef]
  88. Guthery, F.S.; Bingham, R.L. A Primer on Interpreting Regression Models. J. Wildl. Manag. 2007, 71, 684–692. [Google Scholar] [CrossRef]
  89. Shrestha, N. Detecting Multicollinearity in Regression Analysis. Am. J. Appl. Math. Stat. 2020, 8, 39–42. [Google Scholar] [CrossRef]
  90. Wang, Y.M.; Elhag, T.M.S. A Comparison of Neural Network, Evidential Reasoning and Multiple Regression Analysis in Modelling Bridge Risks. Expert Syst. Appl. 2007, 32, 336–348. [Google Scholar] [CrossRef]
  91. Zhou, J.; Li, E.; Wei, H.; Li, C.; Qiao, Q.; Armaghani, D.J. Random Forests and Cubist Algorithms for Predicting Shear Strengths of Rockfill Materials. Appl. Sci. 2019, 9, 1621. [Google Scholar] [CrossRef]
  92. Zhang, Q.; Wu, Z.; Guo, G.; Zhang, H.; Tarolli, P. Explicit the Urban Waterlogging Spatial Variation and Its Driving Factors: The Stepwise Cluster Analysis Model and Hierarchical Partitioning Analysis Approach. Sci. Total Environ. 2021, 763, 143041. [Google Scholar] [CrossRef]
  93. Liu, Y.R.; Li, Y.P.; Sun, J. Statistical Downscaling of Temperature Using Stepwise Cluster Analysis Method—A Case Study in Nur Sultan, Kazakhstan. IOP Conf. Ser. Earth Environ. Sci. 2020, 435, 12019. [Google Scholar] [CrossRef]
  94. Sun, W.; Shi, Q.; Huang, Y.; Lv, Y. Ensemble Learning Enhanced Stepwise Cluster Analysis for River Ice Breakup Date Forecasting. J. Environ. Inf. Lett. 2019, 1, 37–47. [Google Scholar] [CrossRef]
  95. Huang, G.H.; Huang, Y.F.; Wang, G.Q.; Xiao, H.N. Development of a Forecasting System for Supporting Remediation Design and Process Control Based on NAPL-biodegradation Simulation and Stepwise-cluster Analysis. Water Resour. Res. 2006, 42, W06413. [Google Scholar] [CrossRef]
  96. Shokrzadeh, S.; Jozani, M.J.; Bibeau, E. Wind Turbine Power Curve Modeling Using Advanced Parametric and Nonparametric Methods. IEEE Trans. Sustain. Energy. 2014, 5, 1262–1269. [Google Scholar] [CrossRef]
  97. Békés, G.; Kézdi, G. Data Analysis for Business, Economics, and Policy; Cambridge University Press: Cambridge, UK, 2021; p. 738. [Google Scholar]
  98. Sahoo, S.; Jha, M.K. Groundwater-Level Prediction Using Multiple Linear Regression and Artificial Neural Network Techniques: A Comparative Assessment. Hydrogeol. J. 2013, 21, 1865–1887. [Google Scholar] [CrossRef]
  99. Strutz, J.; Martin, J.; Greene, J.; Broadbelt, L.; Tyo, K. Metabolic Kinetic Modeling Provides Insight into Complex Biological Questions, but Hurdles Remain. Curr. Opin. Biotechnol. 2019, 59, 24–30. [Google Scholar] [CrossRef] [PubMed]
  100. Srinivasan, S.; Cluett, W.R.; Mahadevan, R. Constructing Kinetic Models of Metabolism at Genome-scales: A Review. Biotechnol. J. 2015, 10, 1345–1359. [Google Scholar] [CrossRef]
  101. Bellier, N.; Chazarenc, F.; Comeau, Y. Phosphorus Removal from Wastewater by Mineral Apatite. Water Res. 2006, 40, 2965–2971. [Google Scholar] [CrossRef]
  102. Zhang, W.; Ding, Y.; Boyd, S.A.; Teppen, B.J.; Li, H. Sorption and Desorption of Carbamazepine from Water by Smectite Clays. Chemosphere 2010, 81, 954–960. [Google Scholar] [CrossRef]
Figure 1. Quantitative analysis of MSL (Multi-Soil-Layering) publications from 1993 to 1 June 2022. (a) Evolution of published papers, (b) ranking of journals by the number of published papers, (c) most productive countries, (d) most productive researchers.
Figure 1. Quantitative analysis of MSL (Multi-Soil-Layering) publications from 1993 to 1 June 2022. (a) Evolution of published papers, (b) ranking of journals by the number of published papers, (c) most productive countries, (d) most productive researchers.
Water 14 03653 g001
Figure 2. Spatial distribution map of collaborated countries in the MSL field; (a) country collaboration map, (b) country collaboration diagram, (c) Japan collaboration network, (d) China collaboration network, (e) Morocco collaboration network; each collaborative relationship is indicated by a specific color and its importance increases with the thickness of lines connecting two countries.
Figure 2. Spatial distribution map of collaborated countries in the MSL field; (a) country collaboration map, (b) country collaboration diagram, (c) Japan collaboration network, (d) China collaboration network, (e) Morocco collaboration network; each collaborative relationship is indicated by a specific color and its importance increases with the thickness of lines connecting two countries.
Water 14 03653 g002
Figure 3. Co-occurrence analysis map using density visualization. Red circles (hotspots) with large labels reflect the most often occurring items compared to those represented by colored circles ranging from green to blue with small labels. In the co-occurrence map, some extracted items are hidden to avoid overlapping.
Figure 3. Co-occurrence analysis map using density visualization. Red circles (hotspots) with large labels reflect the most often occurring items compared to those represented by colored circles ranging from green to blue with small labels. In the co-occurrence map, some extracted items are hidden to avoid overlapping.
Water 14 03653 g003
Figure 4. Illustration of an experimental model of a multi-soil-layering system with the major removal mechanisms.
Figure 4. Illustration of an experimental model of a multi-soil-layering system with the major removal mechanisms.
Water 14 03653 g004
Figure 5. MSL layers design associated with the removal mechanisms of organic matter and fecal indicator bacteria.
Figure 5. MSL layers design associated with the removal mechanisms of organic matter and fecal indicator bacteria.
Water 14 03653 g005
Figure 6. MSL layers design associated with the removal mechanisms of nitrogen.
Figure 6. MSL layers design associated with the removal mechanisms of nitrogen.
Water 14 03653 g006
Figure 7. MSL layers design associated with the removal mechanisms of phosphorus.
Figure 7. MSL layers design associated with the removal mechanisms of phosphorus.
Water 14 03653 g007
Figure 8. Selections of items in the network map. (a) Aeration, (b) HLR, (c) clogging. The distance between two items reflects the strength of the association, while the lines show the relationships between the items. Each color indicates a cluster that groups the more related terms; aeration, HLR, and clogging belong to the same cluster. To avoid overlapping labels, some extracted items are not presented on the network view.
Figure 8. Selections of items in the network map. (a) Aeration, (b) HLR, (c) clogging. The distance between two items reflects the strength of the association, while the lines show the relationships between the items. Each color indicates a cluster that groups the more related terms; aeration, HLR, and clogging belong to the same cluster. To avoid overlapping labels, some extracted items are not presented on the network view.
Water 14 03653 g008
Figure 9. Performance of single-stage MSL systems while treating a variety of wastewaters at different configurations; scale inside plots indicates the removal efficiency.
Figure 9. Performance of single-stage MSL systems while treating a variety of wastewaters at different configurations; scale inside plots indicates the removal efficiency.
Water 14 03653 g009
Figure 10. The simplified schematization of the multi-soil-layering (MSL) system combined with other technologies (VFCW = vertical flow constructed wetland; SFCW = subsurface flow constructed wetland; SWI = subsurface wastewater infiltration): (a) two-stage vertical flow MSL, (b) topsoil planted MSL with vertical flow MSL, (c) planted vertical flow MSL with SFCW, (d) integrated vertical flow MSL and VFCW, (e) integrated VFCW and MSL with two soil-mixture layers, (f) integrated VFCW and MSL with three soil-mixture layers, (g) trickling filter with horizontal flow MSL, (h) trickling filter with vertical flow MSL, (i) blended carbon sources preceded by aeration system with vertical flow MSL, (j) vertical flow MSL with infiltration section based mixed soil and gravel, (k) vertical flow MSL with sand filter.
Figure 10. The simplified schematization of the multi-soil-layering (MSL) system combined with other technologies (VFCW = vertical flow constructed wetland; SFCW = subsurface flow constructed wetland; SWI = subsurface wastewater infiltration): (a) two-stage vertical flow MSL, (b) topsoil planted MSL with vertical flow MSL, (c) planted vertical flow MSL with SFCW, (d) integrated vertical flow MSL and VFCW, (e) integrated VFCW and MSL with two soil-mixture layers, (f) integrated VFCW and MSL with three soil-mixture layers, (g) trickling filter with horizontal flow MSL, (h) trickling filter with vertical flow MSL, (i) blended carbon sources preceded by aeration system with vertical flow MSL, (j) vertical flow MSL with infiltration section based mixed soil and gravel, (k) vertical flow MSL with sand filter.
Water 14 03653 g010
Figure 11. Pollutant removal efficiency by the reviewed combinations using bar charts.
Figure 11. Pollutant removal efficiency by the reviewed combinations using bar charts.
Water 14 03653 g011
Figure 12. Modeling approaches associated with water quality indicators in the MSL systems with scale inside plot indicates how many times the modeling approach has been utilized.
Figure 12. Modeling approaches associated with water quality indicators in the MSL systems with scale inside plot indicates how many times the modeling approach has been utilized.
Water 14 03653 g012
Table 1. Various combinations of the MSL technology with other systems according to the literature review.
Table 1. Various combinations of the MSL technology with other systems according to the literature review.
Combination UnitsScaleType of WastewaterFlow
(log Units)
Two stage MSL
two vertical flow MSL in
polluted river1000-->71997082-[49]
MSL with CW
MSL + Constructed wetlandpilot scalesolar septic tank effluent743>708072>70>70>702.00[62]
MSL + subsurface flow CWfull scaleagritainment30,000--78--70-[23]
MSL + subsurface flow CWfull scaledomestic5000-9392-7692-[20]
Integrated hybrid systempilot scalerice noodle5080.5-73.260.6-544.80[24]
MSL+ vertical flow CWlab scaleacademic building effluent-84.785.584----[25]
MSL with trickling filter
vertical flow trickling filter +
horizontal flow MSL
lab scalemariculture660--8483---[76]
MSL + iron modified zeolite
trickling filter
Other combinations
MSL + blended carbon sourcespilot scalerural non-point source800--30.3>506460-[9]
MSL + subsurface wastewater
MSL + sand filterslab scaleblack water1009480909354784.50[27]
Table 2. Main advantages and disadvantages of various rural sewage treatment technologies.
Table 2. Main advantages and disadvantages of various rural sewage treatment technologies.
Zeolite (Gravel)
Exert soil filtration, adsorption, and biodegradation functions supported by adequate aeration and HLRSmall land requirement
Low energy
No odors, no insects
Easy operation and maintenance
Adaptation to high pollutant loads
Support high HLR
Moderate sanitary efficiency
Risk of clogging at high HLR
Benefiting from the combined effect of the physical and biochemical properties of soil, artificial media, and microorganismsLow-cost
Low energy
Simple operation
High land requirement
Plants are subject to the effects of the seasons
Low denitrification rates
Periodic maintenance
Odor and insects
Transformation of organic matter into mineral elements that can be assimilated by plantsLow-cost
Minimal energy
Simple operation
Resistance to HLR variations
High sanitary efficiency
High land requirement
Risk of evaporation
High residence time
Odor and insect
Sand filterRocks
Infiltration and purification of wastewater by sand-attached microorganismsLow-cost
Small land requirement
Easy operation and maintenance
High sanitary efficiency
Risk of clogging
Low denitrification rates
Odor and insect
Table 3. Modeling approaches used for water quality indicators in the MSL system.
Table 3. Modeling approaches used for water quality indicators in the MSL system.
ReferencesModeling ApproachInput VariablesOutput VariablesR2ConditionsLimitations
Data-driven models
[16]Neural network (NN)HLR, NH4+, DO, BOD5, and EC (influent)NH4+ (removal)>0.93NN = Large data, input and output variables, activation function, hidden layers and neurons, weight decay, optimizer [84].

Cubist = Large data, input and output variables, committees, instance, pruning, or combining operations [84].

SCA = Large data, input and output variables, continuous and/or discrete variables, nodes, leaf nodes, cutting or merging operations [13,85,86].
QPF = ≥2 input variables, continuous output variable, no multicollinearity [87,88].

MLR = ≥2 input variables, continuous output variable, no multicollinearity [87,88,89].
NN = Difficult to describe connection weights; subjectivity in determining optimal parameters; time consuming; high computational complexity [90].
Cubist = Sensitive to a small dataset; sensitive to the fitness of the dataset; overfitted condition often occurred; high computation time [91].
SCA = High requirements for the predictor; high computational requirements; sensitive to its inputs and internal parameters; usually not well described [92,93,94,95].
QPF = Sensitive to outlier data points; difficulty in interpreting its coefficients; Perform poorly on the predictor’s extremes [96,97].
MLR = Sensitive to outliers data points; fails to capture nonlinear relationship; low performance in large datasets; cannot be used to model data with numerous inputs and outputs; requires numerical values; sensitive to a large number of input variables [84,90,98].
HLR, TKN, TC, DO, and NH4+ (influent)TKN (removal)
HLR, TN, TC, BOD5, and DO (influent)TN (removal)
[5]Stepwise cluster analysis (SCA)Time, DO, ORP, pH, removal of COD, NH3, NO3, and TNsulfamethoxazole removal0.82 (SCA)
0.62 (MLR)
Multiple linear regression (MLR)
[6]Neural network (NN)
SS, pH, EC, DO, and BOD5FC (effluent)0.95 (NN)
0.94 (Cubist)
0.48 (MLR)
Multiple linear regression (MLR)
[11]Neural network (NN)HLR, pH, SS, orthophosphates, TC, TN, TKN of the influentTC (total coliform effluent)0.97 (NN)
0.58 (MLR)
Multiple linear regression (MLR)
[24]Linear modelCOD (influent loading rates)COD (removal/MSL)0.88
COD (effluent/MSL)0.88
NH4+ (influent loading rates)NH4+ (removal/MSL)0.75
NH4+ (effluent/MSL)0.59
[13]Stepwise cluster analysis (SCA)Time, DO, pH, ORP, removal of NH4+, NO3, and TNTN (removal)0.94
Quadratic polynomial function (QPF)PBS, activated sludge, and submerged heightDO, pH, ORP, removal of COD, TP, NH4+, NO3, and TN0.89–0.99
[10]Quadratic polynomial function (QPF)bottom submersion, microbial amendment, aerationremoval of COD, BOD5, TP, TN, NH4+ and NO3 0.77–0.99
[10]Stepwise cluster analysis (SCA)Time, removal of BOD5, TP, TN, NH4+, and NO3COD (removal)0.98
Time, removal of COD, TP, TN, NH4+, and NO3BOD5 (removal)0.95
Time, removal of BOD5, COD, TP, TN, and NO3NH4+ (removal)0.95
Time, removal of BOD5, COD, NH4+, TN, and NO3TP (removal)0.85
Time, removal of BOD5, COD, TP, NH4+, and NO3TN (removal)0.98
Time, removal of BOD5, COD, TP, TN, and NH4+NO3(removal)0.89
[72]Linear regressionpH (effluent)TN (removal)0.32
Kinetic Model
[76]Logistic kinetic model (LKM)Q (salinity)COD (effluent)0.98Kinetic = Kinetic constant, input and output variables, reaction rate coefficient, temperature coefficient [25].Kinetic = Large models result in computational intractability; cannot be used to model data with numerous inputs and outputs; non-linearity; computational tractability; parameter identifiability [99,100].
BOD load, NH4+ load, Iv (media surface area), and T (filter effluent temperature)NH4+ (effluent)0.94
[25]Kinetic ModelInfluent concentration, kinetic coefficientsCOD, BOD5, and NH3 removal≥0.94
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Share and Cite

MDPI and ACS Style

Sbahi, S.; Mandi, L.; Masunaga, T.; Ouazzani, N.; Hejjaj, A. Multi-Soil-Layering, the Emerging Technology for Wastewater Treatment: Review, Bibliometric Analysis, and Future Directions. Water 2022, 14, 3653.

AMA Style

Sbahi S, Mandi L, Masunaga T, Ouazzani N, Hejjaj A. Multi-Soil-Layering, the Emerging Technology for Wastewater Treatment: Review, Bibliometric Analysis, and Future Directions. Water. 2022; 14(22):3653.

Chicago/Turabian Style

Sbahi, Sofyan, Laila Mandi, Tsugiyuki Masunaga, Naaila Ouazzani, and Abdessamad Hejjaj. 2022. "Multi-Soil-Layering, the Emerging Technology for Wastewater Treatment: Review, Bibliometric Analysis, and Future Directions" Water 14, no. 22: 3653.

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop