Small Decentralized Technologies for High-Strength Wastewater Treatment and Reuse in Arid and Semi-Arid Regions

Abstract

Rural and semi-urban areas in arid/semi-arid regions are facing severe water scarcity and a series of environmental challenges nowadays, specifically due to rapid urbanization and economic development, climate change, population growth, increasing water demand, influxes of refugees caused by war and regional political conflict, etc. To solve the emerging problems, the safe reuse of treated wastewater in agriculture can provide an additional water resource for countries with high water scarcity. The aim of this study was to investigate the treatment performance and effectiveness of small decentralized wastewater treatment (DWWT) technologies treating high-strength wastewater with concentrations far beyond the European Union testing ranges of parameters such as five-day biochemical oxygen demand (BOD₅ > 500 mg/L), chemical oxygen demand (COD > 1000 mg/L), or total suspended solids (TSS > 700 mg/L). Four (4) commercially available DWWT technologies with a design capacity of 4–8 PE (population equivalent) were selected and operated with various wastewater compositions in Leipzig, Germany. The technologies were (i) the moving bed biofilm reactor (MBBR), (ii) the sequencing batch reactor (SBR), (iii) the membrane bioreactor (MBR) and (iv) the aerated vertical-flow constructed wetland (AVFCW). This study results clearly demonstrated that the EU-certified small DWWT technologies are quite capable of treating high-strength wastewater and can provide high-quality treated water for safe reuse in rural communities of arid and semi-arid regions. During operation with high-strength wastewater with a mean inflow BOD₅, COD and TSS concentrations of 1532 ± 478, 2547 ± 830 and 546 ± 176 mg/L, a low mean BOD₅ (<10 mg/L), COD (<70 mg/L) and TSS (<15 mg/L) in the outflow of the four systems showed removal efficiency of BOD₅ (>99%), COD (>97%) and TSS (>97%), and met the maximum allowable limit value of water quality class A for reuse in agriculture according to Jordanian and Omani standard. The MBR showed almost a complete removal of Escherichia coli (E. coli) in a range of 6.1–6.9-log removal in the outflow during all three experimental phases and performed best for BOD₅, COD, TSS and pathogen removal when treating high-strength wastewater if properly maintained to prevent potential fouling and clogging of the membrane. Before the final permitting process, long-term monitoring under local temperature and climatic conditions as well as guidelines based on local needs (e.g., in Jordan, Oman, etc.) should be developed to guarantee a minimum level of performance standards of such small DWWT technologies and requirements for operation and maintenance (O&M).

1. Introduction

Water scarcity is one of the main challenges to achieving sustainable development and economic growth, specifically in arid and semi-arid regions worldwide. Many rural and semi-urban areas in arid climates face severe stress on local water resources. Water scarcity is mainly due to rapid changes in climatic conditions, rapid population growth, influxes of refugees caused by war and regional political conflict, increased urbanization, water reallocation from rural to urban regions, rapid socio-economic development and higher standards of living [1,2,3,4,5,6].

Most of the countries in these water-scarce regions (in the Middle East, Gulf countries, North Africa, Southern Europe, etc.) are among the most highly water-stressed countries in the world and they are predicted to face serious shortages of water resources in the future [7,8,9]. Jordan is ranked among the poorest countries in the world in terms of water availability [10]. It has been forecasted that in the next 30 years, water-scarce countries such as Oman and the neighboring regions will run out of water [11]. A recent study demonstrates that 29% of the territory of the European Union (EU), except Italy, was affected by water scarcity conditions between the years 2000 and 2019 for at least one quarter of the year and faced the worst seasonal water scarcity in 2019 [12].

A critical factor for wastewater management in arid and semi-arid countries is the drastic difference in water consumption [13]. As of 2021, the daily water consumption in Europe ranges from 77 to 223 L per capita in selected European countries, where Malta ranked with the lowest daily water consumption of only 77 L per capita and Italy ranked first with the highest water consumption of 223 L per capita [14]. In contrast, the daily water consumption in remote rural areas of Jordan is fairly low (40–80 L per capita) and extremely low (25–50 L per capita) in informal settlements, which results in highly concentrated wastewater [15,16,17]. This is a common issue for wastewater treatment in countries with very limited water supply [13,18]. High water scarcity and hence low per capita consumption of water contributes to extremely high concentrations of organic matter and nutrients in raw wastewater (for example, BOD₅: 600–1500 mg/L, COD: 1000–2500 mg/L, TN: 80–150 mg/L, TP: 20–60 mg/L), which can be generated by public buildings and small-scale operations in rural communities such as schools, university campuses, office buildings, refugee and migrant camps, informal nomadic settlements, military bases, etc. [1,15,19,20,21,22].

In general, high-strength wastewater contains higher soluble organics than typical domestic wastewater [23,24], indicated by its elevated concentration ranges of BOD₅ and COD, fecal matter and nutrients (N and P). Such highly concentrated wastewater is a principal source of environmental pollution and represents vast technical and operational challenges to its treatment [25,26]. Insufficient treatment of such high-strength wastewater can severely reduce the dissolved oxygen (DO) levels in the water if released into the receiving aquatic bodies and may cause severe damage to the aquatic organisms; it can be threatening to the quality of the existing surface water bodies as well as groundwater resources [27,28]. Therefore, it must undergo one or more effective treatment processes before being safe enough to be disposed into any receiving aquatic bodies, such as lakes, rivers, or seas, or to be reused for other purposes (e.g., agricultural irrigation).

In remote rural or peri-urban areas with low income where construction of a centralized wastewater collection and treatment system is not economically feasible, decentralized systems are becoming a popular solution nowadays [29]. Small-scale wastewater treatment systems can be presented as cost-efficient decentralized solutions over centralized treatment systems due to low population densities and scattered settlements in rural areas [30]. High-strength wastewater with very high organic load may potentially impose operational problems in the treatment plants. Therefore, suitable decentralized wastewater treatment (DWWT) technologies need to be selected before successful implementation of treated wastewater reuse on a local scale [13]. Reuse of treated water has been approached as an alternative source of water in arid and semi-arid regions [6].

In principle, small DWWT technologies can vary in size and can include any available technology from simple passive anaerobic systems to technically highly complex solutions [31,32]. Conventional biological wastewater treatment systems include trickling filters (TFs), aerated biofilters (ABs), up-flow anaerobic sludge blankets (UASBs), rotating biological contactors (RBCs), submerged fixed bed reactors (SFBRs), anaerobic baffled reactors (ABRs), membrane bioreactors (MBRs), sequencing batch reactors (SBRs), moving bed biofilm reactors (MBBRs), nature-based solutions (NBSs) such as constructed wetland (CW), etc. [33,34,35]. These processes can effectively remove organic pollutants and nutrients from the wastewater. Challenges for implementing small DWWT technologies in arid and semi-arid regions are the abundance of marketable and internationally available technologies. Therefore, the decision makers face difficulties in assessing the sustainability of the technologies for a given local context.

In the EU, all smaller treatment systems (used for populations up to 50 PE) must be certified according to the European Standard DIN EN 12566-3 [36], which regulates a minimum standard of operation reliability and treatment performance. These state-of-the-art EU-certified smaller treatment plants have already proven their capability in overload and underload conditions in a test facility that has been authorized [37] and tested according to the EU testing range of the conventional wastewater parameter concentrations (BOD₅: 150–500 mg/L, COD: 300–1000 mg/L, TSS: 200–700 mg/L; NH₄-N: 22–80 mg/L and TP: 5–20 mg/L). Nevertheless, these plants are not designed and experimentally tested to treat wastewater beyond these EU testing ranges of the parameters.

High-strength wastewater with high organic and nutrient concentrations beyond the EU testing ranges (for example, BOD₅ > 500 mg/L, COD > 1000 mg/L, NH₄-N > 80 mg/L and TP > 20 mg/L) are not covered by the testing and certification procedures that are currently in place for small DWWT technologies in EU countries. Therefore, existing EU-certified smaller systems operating with high-strength wastewater feeding may not guarantee the fulfillment of the required effluent quality standards that are regulated by the countries in arid and semi-arid regions. Extremely high BOD₅, COD and nutrient concentrations in the influent wastewater may lead to a low treatment performance (i.e., poor effluent quality), as well as unacceptable technical and operational problems such as clogging, excessive sludge production, fouling, etc., and plants become biologically overloaded [1,38].

This study provides an assessment of four (4) small-scale DWWT technologies (4–8 PE) regarding their capability to treat high-strength wastewater effectively and their potential for adaptation to the needs of water-stressed regions. A few previous studies reported on the use of several technologies combined in series or coupled together to improve the treatment performance when treating heavily polluted domestic wastewater [39,40,41]. However, there is a dearth of critical and stand-alone investigation and reliable data on the treatment performance of small DWWT technologies treating high-strength wastewater and operating in parallel for a realistic comparison of their effluents. To our knowledge, no scientific studies have yet addressed and compared the treatment performance and the treated water quality of such smaller systems for agricultural reuse standards.

The aim of this study was to assess the treatment performance, effectiveness and applicability of small DWWT technologies treating high-strength wastewater and the potential reuse of the treated effluent in agriculture. For this, we investigated four (4) commercially available small DWWT technologies (4–8 PE) operating with various wastewater compositions (strengths). The specific objectives of this study were (i) to evaluate the influence of different wastewater compositions on the treatment performance of the selected DWWT technologies; (ii) to compare the effectiveness and applicability of the EU-certified DWWT technologies treating wastewater with extremely high organic and nutrient concentrations that were way beyond the EU testing ranges; and (iii) to compare the treated water quality collected from the effluents of different systems with the maximum allowable limits for agricultural reuse standards regulated in the countries from arid regions (for example, Jordan or Oman) and the EU. Knowledge gaps are highlighted, and suggestions for further research activities on small DWWT technologies are also provided.

2. Materials and Methods

2.1. Site Description

The site that was used to carry out the investigation is located at the Research and Demonstration Centre for decentralized wastewater treatment BDZ e.V. in Leipzig, Germany, where 13 unique small DWWT technologies (design capacity: 4–8 PE) simultaneously run with municipal wastewater under identical climate and operating conditions. The demonstration area (32 m × 17 m) of the BDZ consists of two rows of six (6) demonstration boxes (demo boxes) for small DWWT technologies. Apart from the MBR system (which is found outside of the demo-box area), each demo box has a dimension of 5 m × 7 m, and each of the technologies is carefully installed in those individual demo boxes (Figure 1b).

Except for the NBS i.e., CW, the other demo boxes are covered, and thereby the technologies inside the demo boxes are protected against the weather effect. As backfill material, coarse gravel is used for easier installation and removal. To achieve a high degree of safety in the event of a leakage of the plants inside, a trough made of impermeable concrete is used as a foundation base. The MBR is installed in a container separately from the other 12 demo boxes at the site (Figure 1b).

All treatment systems are connected to the distribution systems for loading, sampling, as well as operation and maintenance from a closed central operating aisle. Wastewater is obtained from a nearby pumping station by using a pipe network and fed on a routine basis into a circulatory system ring line (wastewater pressure line). The ring line is able to supply all the wastewater treatment technologies with the same wastewater simultaneously. The supply via the ring line constantly ensures fresh wastewater for all the technologies installed at this site.

The outflow from the technologies as well as the excess water flow to a collection shaft and are transported back to the discharge shaft via a double pumping station. From there, it is combined with the wastewater from the local network in Leutzsch and goes to the centralized sewage treatment plant at Leipzig-Rosental for further treatment.

Each dosing system comprises a 30 L container, and the feeding takes place via two time-controlled pneumatic gate valves by means of a programmable logic controller (PLC). Feeding each system with wastewater to a freely selectable and country-specific daily flow pattern is possible by simulating different feeding scenarios (e.g., under- or overload conditions, electricity failure, etc.) in order to operate the units under realistic operating conditions.

A separate dosing unit (Figure 1a) can be used to alter the composition of the inflow wastewater. Therefore, a dosing station was built at the site for the preparation and simulation of different wastewater compositions [35]. It consists of a dosing tank with a capacity of 22 m³, a second ring line to feed the technologies with the artificial wastewater, and a container-based mixing tank (800 L). The mixing tank comprises an agitator for rapid mixing of the supplied chemical additives. The existing municipal wastewater can be modified and the concentrations of certain wastewater parameters such as BOD₅, COD, TN, TP, etc. can be spiked with calculated dosing of chemical additives; complete artificial wastewater can also be prepared by using chemicals in supply water at the site. In order to avoid precipitation at the bottom of the dosing tank (22 m³) or any changes of concentrations in the artificially prepared modified wastewater, an integrated agitator continuously operates for homogeneous mixing.

The operation of each DWWT technology with pure municipal wastewater from the local network (via ring line 1) or with artificially prepared wastewater from the dosing tank (using ring line 2) is freely selectable. A separate and fully automatic sample collection chamber with an individual cooling device is installed for each demo box. Representative 24 h mixed samples according to DIN EN 12566-3 [36] and random samples from the outlet and inlet of each DWWT technology can be collected.

2.2. Technology Description

A total of thirteen (13) small-scale DWWT technologies are operating at the site but only four (4) of them were selected for investigation in this study. Three (3) selected EU-certified commercial systems (with design capacity ranges from 4 to 6 PE) supplied by different companies were designated as (i) a moving bed biofilm reactor (MBBR), (ii) a sequencing batch reactor (SBR) and (iii) a membrane bioreactor (MBR). Another one, a nature-based solution (NBS) of 8 PE design capacity, was designated as an aerated vertical-flow constructed wetland (AVFCW) and built according to the German Standard DWA-A 262E [42]. A brief description and treatment process of the selected DWWT technologies are described in the following section.

2.2.1. Moving Bed Biofilm Reactor (MBBR)

One of the main characteristic features of the MBBR system is the movement of media or carrier materials in the fluids. The system incorporates the advantage of an attached growth process where microorganisms grow as a biofilm on the plastic carrier materials [43]. The movement of the support material within the biofilm reactor is achieved either by the agitation produced by aeration (aerobic process) or by mechanical stirrers [43,44]. The MBBR system used in this study essentially consists of three process stages: (1) primary treatment with integrated sludge storage, (2) biological treatment and (3) clarification. All these three process stages take place in an integrated container or tank made of concrete or plastic (single-tank plant). The integrated container contains the following three chambers for those three process stages: the primary treatment tank (1/2 of the container), the biofilm reactor for biological treatment (1/4 of the container) and for and the clarifier for secondary clarification (1/4 of the container).

Wastewater is fed to the first chamber for the pre-treatment (sedimentation stage), which also serves as a sludge storage chamber. The mechanically pre-treated wastewater is then fed to the bioreactor for biological treatment, which is based on the principle of the moving or suspended bed biofilm method. In this bioreactor chamber, a special and high-quality media or carrier material is placed in order to attach the biomass. On these carrier materials, the microorganisms grow by using the organic components of the wastewater as their food and convert it into ecologically harmless by-products. Compressed air is supplied to the biological treatment stage with time-controlled aerators. In order to prevent a wash-out of the carrier materials, this chamber is equipped with a slotted tube. The small treatment plant is equipped with a control unit, from which the compressor and the conveying device of the final settling tank are automatically controlled. To ensure optimum operation with minimum energy consumption, the compressor is operated intermittently (alternating operating and idle periods).

2.2.2. Sequencing Batch Reactor (SBR)

The SBR system used in this study functioned according to the conventional SBR process with different treatment stages or cycles [45]. The wastewater first passes through the preliminary treatment stage, which is connected to the biological stage through an opening in the partition wall. The mechanically treated water from the preliminary treatment stage flows through a submerged overflow baffler into the activation stage. The opening in the partition wall causes the water level in the complete plant to be set at the same level. Thus, the complete surface area of the plant is used as a buffer.

A float valve that is inserted in the activation stage directs the air either to the aeration facility at the bottom of the tank or to the airlift pump located higher up. During low water levels, the activation stage is stirred and aerated by the air input. Following an appropriate wastewater inflow, the valve switches on at a defined water level to the airlift pump. Following the switching off of the valve, the first surge affects the transportation as sludge return flow and is guided via an outlet opening into the buffer tank of the preliminary treatment stage. Then comes the settling phase, where the activated sludge sinks to the bottom and a clarified water zone is formed in the upper region. Following the ending of the settling, the airlift pump is in turn flushed with treated wastewater in several short internal surges. During the clarified water offtake, the outlet opening in the buffer tank is closed and the clarified water can discharge via the clarified water pipe.

The treated wastewater is pumped out until the lower switching point of the float valve is reached. When this is reached, the plant switches on again automatically with mechanical aeration, and a new cycle starts. The cyclic treatment process of the plant is controlled by water level. Depending on the mean water consumption, 1 to 3 cycles can take place per day. As the plant used in this study functions according to the volume of wastewater, even sudden fluctuations in the inflow are treated reliably. Moreover, as the preliminary treatment stage and the SBR chamber are connected, the wastewater has to be transported only once (to the outlet) and thus valuable energy is saved.

2.2.3. Membrane Bioreactor (MBR)

The MBR is a combined system of both biological and filtration processes and is potentially very efficient in treating a wide range of wastewater compositions [23]. The bioreactor acts as a biological treatment process, and the membrane acts as a filter within the filtration process. The MBR used in this study is a two-stage treatment process (max. daily wastewater volume: 0.6 m³/d), which consists of a primary treatment tank and an aeration tank. In the primary treatment tank (which also serves as a wastewater storage tank), an aerated sieve separates biologically degradable coarse matter (e.g., feces and toilet paper) and non-dissolvable substances from the wastewater. A pump then conveys the separated wastewater into the aeration tank, where the organic matter in the wastewater is biologically degraded by microorganisms. An aeration system provides the oxygen needed for this process.

The treated wastewater then passes through micro-filtration membranes with a pore size of 0.4 μm for additional mechanical separation. The pores of the micro-filtration membranes are so small that they retain not only the suspended solids but also bacteria and other microorganisms in the reaction tank and remove them from the effluent. Therefore, the effluent discharged from the MBR system is an absolutely clear, odorless and safe liquid.

The dual-tank MBR system used in this study consists of two double-walled safety tanks (each tank footprint is 0.75 m × 1.15 m) with inlet height > 1.65 m, which can be easily installed in the residential building cellar, a garage or a garden shed, and can be adapted to most decentralized locations. In the standard application, the system is installed within a few hours and ready for immediate use. Due to automatic switching, the biological performance of the system is fully maintained, even during holiday periods of several weeks.

2.2.4. Aerated Vertical-Flow Constructed Wetland (AVFCW)

NBS, like CW, is an environmentally sustainable, socially accepted and cost-effective wastewater treatment technology which shows a strong potential for better secondary effluent treatment with low energy consumption and easy management [46,47,48]. Due to a higher oxygen transfer rate and small footprint, vertical-flow constructed wetlands have been widely applied in wastewater treatment [49,50]. Newly developed VFCWs integrated with active aeration augment the treatment capacity, which leads to a smaller footprint and hence reduces construction costs. It can be used in all types of climates and provide stable treatment performance all year round, and it is less vulnerable to clogging. When the system undergoes continuous aeration, an aerobic environment is created that helps to remove a high amount of organic matter as well as to achieve high nitrification [51].

For this study, an actively aerated vertical-flow constructed wetland (AVFCW) system (area: 8 m², design capacity: 8 PE) was built according to the German standard DWA-A 262E [42] at the site. The construction details were discussed by Rahman et al. [35]. In brief, the system is filled with 90 cm of coarse gravel (8–16 mm) as the main filter media, which is saturated, with a hydraulic retention time (HRT) of 3.5 days. The aeration system consists of irrigation lines on the wetland bottom [52]. Aeration is provided 24 h/d by a fishpond air pump (Aqua Medic, Mistral 4000; design capacity: 66 L/min) connected to the perforated irrigation lines that are evenly placed at the bottom of the wetland bed. The wastewater flow direction is vertical (from the top), and the homogeneous distribution of applied air from the bottom is expected to cause appropriate mixing and saturation with dissolved oxygen (DO) in the filter bed.

A three-chamber settling tank (volume: 2.9 m³) made of concrete serves for the pre-treatment of wastewater, and a mechanical distribution unit (PE tank) containing a submersible pump enables periodic distribution to the wetland. Inflow distribution systems are installed evenly on top of the wetland bed (sub-surface flow) over the horizontal length, and the outflow is collected at the bottom of the tank in a perforated inversed T-shaped pipe connected to an elbow. The elbow discharges the effluent at a level around 5 cm below the gravel surface. The treated wastewater is then sent to a storage tank that is in close proximity to the wetland bed.

The wetland surface was evenly planted with Phragmites australis (at a density of 5 rhizomes/m²). Before the experimental phases were started, the plants were allowed to grow and the biofilms to develop within the rhizosphere. During this acclimatization period of 8–10 weeks, the wetland was fed with a small amount of pre-treated wastewater from time to time and aerated continuously.

2.3. Modified Wastewater Preparation

Instead of preparing entirely artificial wastewater by using drinking water, the already existing municipal wastewater at the site was modified and the concentrations spiked by calculated dosing of several chemical additives. It was adjusted to target the concentration of certain parameters such as BOD₅, COD, TN, TP, TSS, etc. within the inflow wastewater in different experimental phases during this study.

Maisonnave et al. [53] showed the composition of artificial wastewater representing typical domestic wastewater (in mg/L): BOD₅, 400; COD, 880; SS, 350; TKN-N, 80; P, 10. The concentrations of the ingredients that were used (in mg/L) were as follows: glycine, 250; glycerol, 300; K₂HPO₄: 10; NH₄Cl, 60; Na₂HPO₄, 30. Based on the study by Maisonnave et al. [53], the composition of the ingredients was adapted and simulated with already existing municipal wastewater at the site. In general, the concentrations of the existing wastewater parameters were increased by adding organic carbon sources (glycine and glycerol) and nutrient sources (NH₄Cl, K₂HPO₄, Na₂HPO₄), and thereby modified wastewater of defined strength was prepared in this study.

After rapid and homogeneous mixing of the added chemicals in the mixing tank, the thick solution from the mixing tank was pumped to the dosing tank or wastewater storage tank (22 m³) (Figure 1a). There, it was mixed with the already existing municipal wastewater of a specified volume, and the concentrations of the specific wastewater parameters (BOD₅, N, P, etc.) were increased. All the ingredients used in the preparation of the modified wastewater for spiking the concentrations were dissolved almost exclusively in water. To increase the concentrations of filterable substances or TSS, a pre-calculated amount of thick digested sludge from a nearby sludge storage tank was pumped directly to the dosing tank and added to the volume of existing wastewater.

After every filling and proper mixing of the freshly prepared modified wastewater, a sample was taken from the dosing tank using ring line 2 to assess the concentrations of typical wastewater parameters and thereby know the strength of the inflow wastewater that is fed to the selected technologies.

2.4. Operational Conditions

Based on low water consumption and hence low wastewater generation or discharge (volume) per capita in a day within the arid and semi-arid regions, wastewater compositions of three different strengths were simulated and prepared in three experimental phases (I, II and III) of this study. The same mean influent BOD₅, TN and TP load (in grams per capita and day) that are typical of the rural areas in arid regions were used and remained constant in the three experimental phases of this study. Depending on the water consumption on a daily basis, the concentrations of these parameters (BOD₅, TN and TP) vary, i.e., they either dilute when more volume of water is consumed and thereby more volume of wastewater is discharged or become strong when less volume of water is consumed and therefore less volume of wastewater is discharged. The wastewater discharge rates were incrementally reduced and spiked with chemical additives in order to increase the concentrations of these specific parameters in this study. Due to significant logistic issues, it was not possible to carry out this experiment in any rural areas of arid regions with extreme water scarcity (e.g., in Jordan), and therefore municipal wastewater at the site in Germany was used in a simulation to achieve the target concentrations of these specific parameters. In this study, the municipal wastewater discharge rate on a daily basis was assumed as 150 L per capita at the site [42]. Three mean BOD₅ concentrations were targeted as 300, 600 and 1200 mg/L in three experimental phases in Phase I, II and III, respectively. Similarly, three mean concentrations of 60, 120 and 240 mg/L TN and 10, 20 and 40 mg/L TP were targeted to be achieved within the inflow wastewater in experimental Phases I, II and III, respectively (

Table 1. Mean influent BOD₅, TN, TP load and wastewater discharge volume used to simulate the targeted mean concentrations of BOD₅, TN and TP in three experimental phases (Phases I–III) in this study.

Experimental Phase	Wastewater Type	Wastewater Discharge [L/(PE × d)]	BOD5 Load [g/(PE × d)] c	TN Load [g/(PE × d)] c	TP Load [g/(PE × d)] c	BOD5 (mg/L)	TN (mg/L)	TP (mg/L)
Phase I	real a	150	45	9	1.5	300	60	10
Phase II	modified b	75	45	9	1.5	600	120	20
Phase III	modified b	37.5	45	9	1.5	1200	240	40

Notes: ^a existing municipal wastewater at the site; ^b existing wastewater mixed with a calculated amount of chemical additives; ^c adapted from Bertanza and Boiocchi [54].

).

For the EU certification process (including a practical test in a test field) of small DWWT technologies, the concentration range of the raw domestic wastewater parameters is already defined according to DIN EN 12566-3 [36]. The concentration ranges of the parameters for the EU practical testing and actual concentration ranges in the influent wastewater (strengths) in three experimental phases (Phases I–III) are shown in

Table 2. The concentration ranges of the parameters for the EU testing and in the influent wastewater (strengths) used within the three experimental phases (Phases I–III).

Parameters	Unit	Concentration (Range)
		EU Practical Test DIN EN 12566-3 [34]	Phase I a (Real Wastewater)	Phase II b (Modified Wastewater)	Phase III c (Modified Wastewater)
BOD5	mg/L	150–500	250–400	500–800	1000–1900
COD	mg/L	300–1000	460–980	980–1550	1550–3400
TSS	mg/L	200–700	100–280	200–450	400–750
NH4-N	mg/L	22–80	20–80	60–130	140–420
TP	mg/L	5–20	3–10	10–25	25–60

Notes: ^a number of samples, n = 11; ^b n = 12; ^c n = 12.

.

All four selected DWWT technologies were fed by the wastewater of the same influent strength as mentioned in each experimental phase and operated with the same adjusted daily flow pattern. For each cycle of feeding, the influent wastewater was fed to each system in intermittent mode with a fixed volume and time using an automatic time-controlled PLC.

Table 3. Inflow feeding rates of the four selected systems operated during three experimental phases (Phases I–III) in this study.

System	Design Capacity (PE)	Influent Flow Rates (L/d)
		Phase I (Real Wastewater)	Phase II (Modified Wastewater)	Phase III (Modified Wastewater)
MBBR	4	600	300	150
SBR	6	900	450	225
MBR	4	600	300	150
AVFCW	8	1200	600	300

shows the influent flow rates of the four selected DWWT technologies operated in three experimental phases in Phases I, II and III.

To avoid freezing in the dosing tank (22 m³), which operates under outside air temperature at the site, the whole investigation and operation of the selected technologies was paused for 3 weeks in winter when the outside air temperature was extremely low (below 0 °C). The duration of the whole investigation with three experimental phases was one year and the mean air temperature was measured in a wide range of 10–35 °C within the course of the study period at the site.

2.5. Sampling and Analysis

Throughout all three experimental phases (Phases I–III), the samples from the influent and the effluents of the four respective DWWT technologies were analyzed on a weekly routine basis for measuring the concentrations of the following parameters: BOD₅ (DIN 38409 H52, WTW OxiTOP^®, manufactured by Xylem Analytics, Weilheim, Germany), COD (TNTplus^TM 821/822, HR 20–1500 mg/L COD manufactured by HACH^®, Düsseldorf, Germany), ammonium–-nitrogen (NH₄-N) [55], nitrate–nitrogen (NO₃-N) [56], total nitrogen (TN) (TNTplus™ Vial Test TNT828, UHR 20-100 mg/L N, manufactured by HACH^®, Düsseldorf, Germany), total phosphorous (TP) [57] and TSS (by using vacuum filtration). The spectrophotometer DR 2800 (Hach Lange, Duesseldorf, Germany) was used for reading the concentrations of COD and TN, according to the standard method for COD and TN tests specified by the manufacturer (HACH^®, Düsseldorf, Germany). Dissolved oxygen (DO) (ConOx^®, WTW Germany), water temperature (T °C) and pH electrode (SenTix^® pH) using a handheld meter (Multi 350i^®, WTW, Weilheim, Germany), redox potential (E_h) (electrode SenTix^® ORP, WTW, Weilheim, Germany), electrical conductivity (EC) (conductivity meter Cond 330i, WTW, Weilheim, Germany) of the inflow and outflow samples were measured and E. coli ([58] were quantified simultaneously by using Colilert-18 Quanti-Tray^TM method (IDEXX, Westbrook, ME, USA) to investigate treatment performances of the technologies and to assess the requirements for system maintenance.

Pollutant removal efficiency was calculated as concentration reduction or removal in percentage using the following Equation (1):

$$R=\frac{\left(c_{in}-c_{out}\right)}{c_{in}}\times 100$$

(1)

where R is the removal efficiency (%), c_in is the influent concentration [mg/L] and c_out is the effluent concentration [mg/L]. Based on available data from all 4 systems, we calculated and compared the pollutant removal efficiencies only with the concentrations. Random samples were taken from the systems on a weekly basis, and the removal efficiency was calculated as the weekly average in this study.

2.6. Data and Statistical Analysis

The concentrations of different parameters in the samples from the influent and effluent were reported in the format of mean ± SD in this study. Microsoft Excel 2013 software (product name and activation status: Microsoft Office Professional Plus 2013; version: Microsoft^® Excel^® 2013 (15.0.5571.1000) MSO (15.0.5571.1000) 32-bit) package was used to determine the mean and standard deviation of the sample concentrations. Statistical analysis was performed using a one-way analysis of variance (ANOVA) test with a 95% significance level of difference within Microsoft Excel 2013 to compare the treatment performance of the four selected DWWT technologies in terms of differences between mean concentrations at the influent and effluent, as well as mean removal efficiencies with various wastewater compositions or strengths. The statistical test results and differences were considered significant at p-values of less than 0.05 (p < 0.05).

3. Results and Discussion

3.1. Treatment Performance Assessment of the Four DWWT Technologies

Mean concentrations of different physico-chemical parameters and microbial quantifications at the inlet and outlet and thereby removal efficiencies of the four DWWT technologies during the whole experimental period are discussed in the following sections. Water quality parameters were statistically analyzed to determine any significant changes in the treatment performance of the four DWWT technologies when the influent wastewater strength was increased stepwise in the three experimental phases (Phases I–III).

Treated wastewater qualities from the outflow of four selected systems achieved in three experimental phases are also compared with the maximum allowable limit values for reuse in agriculture that are regulated in two arid climate countries Jordan and Oman. Both the Jordanian and Omani standards are also compared with the strictest EU standard for reclaimed water quality requirements in agricultural reuse as shown in

Table 4. Allowable limit values and different classes of treated wastewater for reuse in Jordan according to Jordanian national standard JS 893/2006 adapted from Abdallat et al. [59]; maximum quality limits in Oman according to Oman Ministerial Decision 145/1993 [60]; and minimum reclaimed water quality requirements for agricultural irrigation in the EU [61] (n.p.: not provided).

Parameter	Unit	Allowable Limits for Reuse in Jordan			Maximum Wastewater Quality Limits in Oman		Minimum Reclaimed Water Quality Requirements According to the EU
		Class A a	Class B b	Class C c	Class A d	Class B e	Class A f	Class B g	Class C h	Class D i
BOD5	mg/L	30	200	300	15	20	10	25	25	25
COD	mg/L	100	500	500	150	200	n.p.	n.p.	n.p.	n.p.
TSS	mg/L	50	150	150	15	30	10	35	35	35
DO	mg/L	>2	n.p.	n.p.	n.p.	n.p.	n.p.	n.p.	n.p.	n.p.
pH	–	6–9	6–9	6–9	6–9	6–9	n.p.	n.p.	n.p.	n.p.
EC	μS/cm	n.p.	n.p.	n.p.	2000	2700	n.p.	n.p.	n.p.	n.p.
NH4-N	mg/L	n.p.	n.p.	n.p.	5	10	n.p.	n.p.	n.p.	n.p.
NO3-N	mg/L	30	45	45	50	50	n.p.	n.p.	n.p.	n.p.
TN	mg/L	45	70	70	n.p.	n.p.	n.p.	n.p.	n.p.	n.p.
TP	mg/L	n.p.	n.p.	n.p.	30	30	n.p.	n.p.	n.p.	n.p.
E. coli	MPN/100 mL (log10 E.coli)	100 (2.0)	1000 (3.0)	n.p.	200 (2.3)	1000 (3.0)	10 (1.0)	100 (2.0)	1000 (3.0)	10,000 (4.0)

^a Cooked vegetables, parks and playgrounds within city limits (restricted irrigation); ^b fruit trees, sides of roads outside city limits and landscape (restricted irrigation); ^c field crops, industrial crops and forest trees; ^d vegetables likely to be eaten raw, fruit likely to be eaten raw and within two (2) weeks of any irrigation, public parks, hotel lawns and recreational areas with public access; ^e vegetables to be cooked or processed, fruit if no irrigation within two (2) weeks of cropping, fodder, cereal seed crops, pasture, no public access; ^f all food crops, including root crops consumed raw and food crops where the edible portion is in direct contact with reclaimed water (all irrigation methods allowed); ^g food crops consumed raw where the edible portion is produced above ground and is not in direct contact with reclaimed water and processed food crops and nonfood crops, including crops to feed dairy or meat-producing animals (all irrigation methods allowed); ^h food crops consumed raw where the edible portion is produced above ground and is not in direct contact with reclaimed water, processed food crops and nonfood crops including crops to feed dairy or meat-producing animals (drip irrigation or other irrigation methods); ⁱ industrial and energy crops and seeded crops (all irrigation methods allowed).

.

3.1.1. pH, E_h, DO, EC and T

Table 5. Measured mean values of pH, E_h, EC, T and DO concentrations in the inflow and outflow of the four selected systems with different wastewater compositions in the three experimental phases (Phases I–III).

Experimental Phase	Sampling Point/System	pH (-)	Eh (mV)	DO (mg/L)	EC (µS/cm)	T (°C)	Number of Samples
I	Inflow	7.7 ± 0.4	−100 ± 53	0.4 ± 0.2	1240 ± 550	11.3 ± 2.6	8
	Outflow
	MBBR	7.4 ± 0.2	295 ± 130	9.4 ± 1.3	1062 ± 438	9.9 ± 2.9	8
	SBR	7.2 ± 0.3	299 ± 116	10.7 ± 0.8	1090 ± 442	9.9 ± 2.7	8
	MBR	7.0 ± 0.3	294 ± 130	5.8 ± 2.3	1044 ± 440	13.8 ± 2.0	8
	AVFCW	6.9 ± 0.5	286 ± 103	10.5 ± 1.5	1116 ± 455	8.2 ± 3.5	8
II	Inflow	7.0 ± 0.5	−224 ± 61	1.1 ± 1.0	1815 ± 281	19.6 ± 3.1	12
	Outflow
	MBBR	6.8 ± 0.4	180 ± 25	8.2 ± 1.4	1377 ± 268	17.1 ± 4.3	12
	SBR	6.6 ± 0.7	186 ± 22	7.8 ± 1.8	1300 ± 184	17.7 ± 3.7	12
	MBR	7.8 ± 0.3	98 ± 92	3.9 ± 2.2	1478 ± 182	19.9 ± 2.4	12
	AVFCW	7.2 ± 0.4	180 ± 32	7.7 ± 2.1	1276 ± 292	13.2 ± 4.8	12
III	Inflow	7.2 ± 0.5	−227 ± 49	0.2 ± 0.1	3474 ± 858	20.3 ± 4.1	12
	Outflow
	MBBR	6.9 ± 0.8	209 ± 22	7.3 ± 3.1	1915 ± 427	16.1 ± 3.7	12
	SBR	5.8 ± 0.5	222 ± 29	9.1 ± 1.1	2170 ± 523	15.5 ± 3.4	12
	MBR	5.9 ± 1.0	243 ± 33	8.0 ± 1.4	2263 ± 831	18.4 ± 3.3	12
	AVFCW	6.2 ± 0.5	230 ± 46	8.6 ± 2.8	2599 ± 833	11.6 ± 3.8	12

shows the mean inflow and outflow values of pH, E_h, T and DO concentrations in the four selected DWWT technologies that were observed during the three experimental phases of this study.

The statistical analysis showed significant (p < 0.05) differences in pH, E_h, EC and T values and DO concentrations between the three influent wastewater compositions used in the three experimental phases from Phase I to Phase III in this study.

The pH of the inflow wastewater in all three experimental phases was in the range of 5.3–8.4 with a mean value of 7.7 ± 0.4, 7.0 ± 0.5 and 7.2 ± 0.5 in Phases I, II and II, respectively. In general, the mean pH value was significantly reduced (p < 0.05) in the outflow of all four systems as compared to the inflow in all three experimental phases. During operation with high-strength wastewater in Phase III with a mean pH value of 7.2 ± 0.5 in the inflow, a significant reduction in mean pH (p < 0.05) in the range of 5.8–6.9 was observed in the outflow of all four selected systems. As compared to Jordanian and Omani standards of pH in a range of 6.0–9.0 for all classes of treated water, only the SBR and MBR systems showed a mean pH of <6.0 in the outlet and did not meet the standard when operated with high-strength wastewater in Phase III. Prior to the reuse of such treated water with a slightly low pH, some pH adjustment may potentially be needed.

Low redox potential in a range of −340 to −31 mV was recorded in the inflow samples from the dosing tank during all three experimental phases in this study. A general trend of increasing the mean redox value in the range from 98 to 299 mV in the outflow of the four systems suggested that aerobic conditions prevailed within all the systems (Table 5). The lowest mean value of −227 ± 49 mV was recorded in the inflow samples of high-strength modified wastewater in Phase III, as compared to the other two phases (Phases I and II). During the operation with high-strength wastewater in Phase III with a mean redox potential of −227 ± 51 mV in the inflow, significantly higher (p < 0.05) mean redox values were recorded in the range of 209–243 mV in the outflow of the four systems. The observed increase in redox potential at the outlet of all the systems was perhaps due to an overall decrease in organic load [58]. A comparatively low mean redox potential of 98 ± 92 mV was observed in the outflow of the MBR system in Phase II.

As expected, very low mean DO concentrations in the range of 0.2–1.1 mg/L were analyzed in the inflow samples of the three experimental phases. During operation with different wastewater compositions in Phases I–III, mean DO concentrations increased substantially in a range of 7.3–10.7 mg/L in the outlet of the MBBR, SBR and AVFCW systems. High DO concentration at the outlet clearly showed an abundance of oxygen within these systems and potentially favored redox conditions necessary for the oxidation of the wastewater pollutants. Only the MBR showed a comparatively low DO of 5.8 ± 2.3 mg/L and 3.9 ± 2.2 mg/L in the outflow in Phase I and Phase II, respectively. After the maintenance of the MBR system, it showed relatively higher DO with a mean concentration of 8.0 ± 1.5 mg/L in the outflow in Phase III, as compared to the previous two experimental phases (Phases I and II). However, mean DO concentrations at the outlet of all four systems met the allowable limit concentration of treated water quality class A for agricultural reuse recommended by the Jordanian standard, which is >2.0 mg/L (Table 4).

Mean EC in the inflow wastewater was significantly increased (p < 0.05) stepwise and recorded as 1240 ± 550, 1815 ± 281 and 3474 ± 858 µS/cm in experimental Phases I, II and III, respectively (Table 5). A general trend of decreasing the mean value of EC within the outflow of the four systems was observed. During the operation with high-strength wastewater in Phase III, mean EC showed significant changes (p < 0.05) in the inflow and outflow of all systems. EC is a common indicator of salinity and only the MBBR system with a mean EC of 1915 ± 427 µS/cm in the outflow met class A water quality (<2000 µS/cm) according to the Omani standard when treating high-strength wastewater in this study. The effluent from an MBBR system treating laundry wastewater also exhibited high EC in the range of 1800–2000 µS/cm [62]. Mean ECs recorded in the outlet of the other three systems were in compliance with the class B water quality limit (<2700 µS/cm). To protect the soil from high salinity, direct reuse of the treated effluents from the four systems with a comparatively high EC may not be possible, and therefore innovative irrigation methods (e.g., sub-surface irrigation) can be suggested.

The mean temperature of the inflow wastewater was recorded as 11.3 ± 2.6, 19.6 ± 3.1 and 20.3 ± 4.1 °C during the experimental Phases I, II and III, respectively (Table 5). In general, the mean temperature in the outflow of the MBBR, SBR and AVFCW systems decreased slightly in comparison with the mean inflow temperature. Comparatively low water temperatures in the range of 8.2–13.2 °C were observed in the outflow of the AVFCW system during the three phases (Phases I–III). The mean water temperature increased slightly in the outflow of the MBR system in Phases I and II but decreased in Phase III as compared to the mean inflow water temperatures.

3.1.2. Removal of BOD₅ and COD

Treatment performance with respect to pollutant removal efficiency and the concentration profile of BOD₅ and COD from the inflow and outflow of the four selected DWWT technologies during the three experimental phases are shown in

Table 6. Treatment performance with different wastewater compositions observed by analyzing the mean concentrations of the conventional wastewater parameters at the inflow and outflow of the four selected DWWT technologies in the three experimental phases (Phases I–III).

Experimental Phase	Sampling Point/ System	BOD5 (mg/L)	COD (mg/L)	TSS (mg/L)	NH4-N (mg/L)	NO3-N (mg/L)	TN (mg/L)	TP (mg/L)	log10 E. coli (MPN/100 mL)	Number of Samples
I	Inflow	292 ± 60	608 ± 152	207 ± 64	44.2 ± 19	0.5 ± 0.1	60 ± 19	7.9 ± 3	7.1 ± 0.3	11
	Outflow
	MBBR	4 ± 3 (99 a)	41 ± 9 (93 a)	7 ± 3 (97 a)	2.8 ± 2 (94 a)	21 ± 4	28 ± 7 (54 a)	6.2 ± 1 (22 a)	4.8 ± 0.5 (2.3 b)	10
	SBR	4 ± 3 (99)	35 ± 14 (94)	6 ± 5 (97)	0.6 ± 0.5 (99)	33 ± 9	38 ± 9 (37)	6.6 ± 2 (17)	3.9 ± 0.4 (3.2)	10
	MBR	1 ± 0.8 (99)	20 ± 4 (97)	1.2 ± 1 (99)	0.8 ± 0.7 (98)	27 ± 10	29 ± 12 (52)	5.4 ± 2 (32)	0.2 ± 0.1 (6.9)	10
	AVFCW	2 ± 1.7 (99)	34 ± 6 (94)	3 ± 1 (99)	3.9 ± 3.8 (91)	40 ± 9	44 ± 9 (27)	4.2 ± 1 (48)	4.7 ± 0.6 (2.4)	11
II	Inflow	664 ± 107	1288 ± 277	310 ± 85	88.4 ± 27	0.5 ± 0.1	123 ± 23	18.8 ± 5	6.9 ± 0.4	12
	Outflow
	MBBR	5 ± 3 (99 a)	41 ± 6 (97 a)	4 ± 1 (99 a)	3.8 ± 3 (96 a)	41 ± 7	47 ± 12 (62 a)	13 ± 3 (29 a)	3.6 ± 0.6 (3.3 b)	12
	SBR	7 ± 6 (99)	57 ± 23 (96)	13 ± 9 (96)	1.1 ± 0.7 (99)	43 ± 20	51 ± 23 (59)	14 ± 3 (27)	4.1 ± 1.0 (2.8)	12
	MBR	4 ± 3 (99)	33 ± 14 (97)	2.1 ± 2 (99)	49 ± 38 (45)	3.8 ± 3	46 ± 32 (63)	8 ± 6 (58)	0.1 ± 0.1 (6.8)	12
	AVFCW	3 ± 2 (99)	36 ± 4 (97)	3 ± 1 (99)	1.0 ± 0.5 (99)	31 ± 14	33 ± 13 (73)	11 ± 2 (44)	4.2 ± 0.8 (2.7)	12
III	Inflow	1532 ± 478	2547 ± 830	546 ± 176	267 ± 110	0.5 ± 0.1	367 ± 133	45.3 ± 12	6.2 ± 0.6	12
	Outflow
	MBBR	10 ± 9 (99 a)	70 ± 39 (97 a)	13 ± 9 (98 a)	33 ± 30 (87 a)	46 ± 13	89 ± 35 (76 a)	20 ± 3 (56 a)	3.8 ± 0.6 (2.4 b)	12
	SBR	4 ± 2 (99)	70 ± 30 (97)	15 ± 7 (97)	53 ± 38 (80)	83 ± 34	190 ± 71 (48)	24 ± 5 (48)	2.8 ± 0.8 (3.4)	11
	MBR	9 ± 7 (99)	62 ± 46 (98)	4.5 ± 2 (99)	60 ± 36 (78)	66 ± 31	143 ± 59 (61)	37 ± 15 (18)	0.1 ± 0.1 (6.1)	12
	AVFCW	6 ± 5 (99)	64 ± 18 (97)	8 ± 3 (98)	46 ± 33 (83)	101 ± 33	166 ± 75 (55)	14 ± 5 (69)	3.6 ± 0.5 (2.6)	11

^a mean percentage removal efficiency in all the following parentheses; ^b log-removal in all the following parentheses.

and Figure 2.

The results of a very low mean BOD₅ in the outflow of all four systems showed highly efficient removal efficiency (>99%) in all three experimental phases in this study (Table 6). Significant reduction (p < 0.05) of BOD₅ concentrations was observed at the outlet of all systems as compared to the respective inflow in three phases. In general, BOD₅ concentration dynamics in the outflow of four systems showed no significant changes (p < 0.05) with different wastewater strengths as inflow (Figure 2). No potential impact of increased wastewater strength was also observed on BOD₅ removal performance. Only a fluctuating tendency in the outflow BOD₅ concentrations from the systems was observed (Figure 2). During operation with high-strength wastewater in experimental Phase III with a mean BOD₅ concentration of 1532 ± 478 mg/L in the inflow, the mean BOD₅ concentration of <10 mg/L in the outflow of the four systems accounted for a highly efficient BOD₅ removal (>99%) and successfully met the maximum allowable limit value of class A water quality requirements suggested by the Jordanian standard for restricted irrigation (<30 mg/L), the Omani standard (<15 mg/L) and the EU regulations (<10 mg/L) for reuse in agricultural irrigation purposes.

As can be seen in Figure 2, the outflow COD concentrations from the systems slightly raised and fluctuated with increasing COD concentrations in the inflow from Phase I to Phase III. However, the overall mean COD concentrations in the inflow were significantly decreased (p < 0.05) in the outflow from all four systems, with an average removal efficiency in the range of 93–98% in all three experimental phases. During Phase III with high-strength wastewater as inflow with a mean COD concentration of 2547 ± 830 mg/L, the mean COD concentrations in the outflow from all systems resulted in a highly efficient mean COD removal of >97% and were much lower than the maximum allowable COD limit value for class A water quality according to Jordanian standards (<100 mg/L) as well as Omani standards (<150 mg/L). No COD standard of reclaimed water quality for agricultural irrigation purposes is suggested by the EU. High-strength wastewater is rich in organic matter, which is a very good source of carbon and nutrients (N and P) for the microbes to produce energy and synthesize chemical products [63], and hence highly efficient COD removal was observed in this study.

The BOD₅ and COD from high-strength wastewater can be degraded by both aerobic and anaerobic bacteria that can be potentially grown within the four DWWT technologies, and aerobic degradation with a high DO concentration is also fast and important [49]. Limited oxygen supply restricts COD degradation for wastewater treatment [64], and in this study, the mean DO concentrations in the range of 7.3–9.1 mg/L in the outlet of the systems were observed in experimental Phase III operating with high-strength wastewater (Table 5). These relatively high DO concentrations inside the four selected systems were the potential reason for highly efficient BOD₅ and COD removal. Zhao et al. [65] showed that with the increasing influent wastewater strength, the limitation of oxygen became more obvious and resulted in decreasing COD removal efficiencies in vertical flow constructed wetlands (VFCWs) without aeration. However, in this study, the air supplied via aeration facilities within the systems provided the oxygen needed to degrade the organic matter with increasing strength of inflow wastewater, and therefore BOD₅ and COD degradation were never restricted due to a limited supply of oxygen.

3.1.3. TSS Removal

The TSS concentration profile from the inflow and outflow of the four systems with various wastewater compositions in three experimental phases is shown in Figure 3.

The dynamics of outflow TSS concentrations from all systems showed a fluctuating trend in all experimental phases as compared to the inflow. The SBR system showed a slight rise of TSS concentration in the outflow when the system was fed with more concentrated wastewater in Phases II and III. However, the mean inflow TSS concentrations of 207 ± 64, 310 ± 85 and 546 ± 176 mg/L in three experimental Phases I, II and III were reduced to mean outflow TSS concentrations in the range of 1.2–15 mg/L, which accounted for highly efficient mean TSS removal of >96% in all four systems in this study (Table 6). Therefore, all four systems proved to be highly effective at removing suspended solids from high-strength wastewater with a very high TSS concentration and met the treated water quality class A for TSS according to Jordanian (<50 mg/L) and Omani standards (<15 mg/L). However, a few measurements with TSS > 20 mg/L in the outflow of the MBBR and SBR systems exceeded class A water quality standard (<15 mg/L) and met class B standard (<30 mg/L) according to Omani regulations (Figure 3). The MBR and AVFCW systems showed highly efficient TSS removal and the mean outflow TSS concentrations in all three experimental phases even met the strictest EU standard for class A water quality (<10 mg/L) recommended for reuse in agriculture (Table 4 and Table 6). A similar trend in the effluent TSS concentrations (<15 mg/L) for a pilot-scale recirculating vertical flow constructed wetland was reported by Sklarz et al. [66].

3.1.4. NH₄-N, NO₃-N and TN Removal

The treatment performance of NH₄-N, NO₃-N and TP and their dynamics in the inflow and outflow of the four DWWT technologies during the three experimental phases are shown in Table 6 and Figure 4, respectively.

The four systems were almost completely nitrified by converting ammonia–nitrogen (NH₄-N) to nitrate–nitrogen (NO₃-N) in experimental Phase I and also in Phase II, except for the MBR system, which required sufficient maintenance. In general, the results demonstrated enhanced nitrification and therefore highly efficient removal of NH₄-N in the range of 91–99% in experimental Phases I and II in this study. Mean concentrations of NH₄-N within the outflow also achieved class A water quality standard for NH₄-N (<5 mg/L) according to Omani guidelines for reuse. The only exception was the MBR system in Phase II, where a high NH₄-N mean concentration of 49 ± 38 mg/L in the outflow demonstrated a mean NH₄-N removal efficiency of 45% only. Low DO with a mean concentration of 3.9 ± 2.2 mg/L in the outlet of the MBR system during this phase was potentially associated with a low nitrification rate [67], and therefore only a low NO₃-N mean concentration of 3.8 ± 3 mg/L was observed (Table 6). This was probably due to the insufficient maintenance of the MBR system and clogging of the membrane during this phase. After changing the membrane and some necessary maintenance work needed for the MBR system during this phase (Phase II), the outflow NH₄-N concentration decreased remarkably and showed better NH₄-N removal again (Figure 4).

When operating with high-strength wastewater in experimental Phase III, relatively high NH₄-N mean concentrations in the outflow suggested that there was room for optimizing NH₄-N removal performance within the systems. In experimental Phase III with a very high NH₄-N concentration in the inflow, the dynamics of NH₄-N concentration showed a sharp rise in the outflow of all four systems (Figure 4). A mean inflow NH₄-N concentration of 267 ± 110 mg/L was attributed to a mean outflow concentration in the range of 33–60 mg/L, which resulted in a NH₄-N removal efficiency with a great extent and showed in the range of 78–87% within the four systems. (Table 6). However, a slightly lower mean NH₄-N removal (78–87%) and at the same time, a highly efficient mean COD removal of >97% in all four systems in experimental Phase III was consistent with the results achieved by Li et al. [68]. Previous results indicated that artificial aeration increased the DO concentrations in the AVFCW treating heavily polluted water, which significantly favored the removal of organic matter and NH₄-N [51].

The main removal process of NH₄-N is nitrification, where NH₄-N is converted into NO₃-N. This process needs high DO concentration [48] and afterward, this NO₃-N is transformed into N₂O or N₂ via a denitrification process in order to be permanently removed from the system [69]. The four selected DWWT technologies showed higher mean DO concentration in the outflow (7.3–9.1 mg/L) during operation with high-strength wastewater in this investigation. A relatively higher NO₃-N with a mean concentration in the range of 46–101 mg/L was observed in the four systems in the experimental phase, and this NO₃-N was shown as the predominant form of nitrogen in the effluent in experimental Phase III with high-strength wastewater (Table 6). Only the MBBR system outflow mean NO₃-N concentration (46 ± 13 mg/L) achieved the treated water quality of Omani class A standard (<50 mg/L) for reuse in agriculture. The reason for a comparatively low mean NO₃-N concentration in the effluent might be due to the fact that the MBBR with a thick biofilm potentially allows for some anoxic environment that promotes denitrification. The mean outflow concentrations of the other three systems did not comply with the allowable limit suggested by the Omani standard. Nitrification usually becomes the limiting step for nitrogen removal in conventional VFCWs due to insufficient DO [70]. However, in this study, the AVFCW system with active aeration provided sufficient DO (8.6 ± 2.8 mg/L), and this was also in agreement with other artificially aerated constructed wetlands with saturated DO levels in the range of 8–11 mg/L [71]. Therefore, it is well documented that all four technologies in this study were sufficiently aerated within the systems.

Mean inflow TN concentrations of 60 ± 19, 123 ± 23 and 367 ± 133 mg/L were reduced in the outflow, which produced a mean TN removal efficiency in the range of 27–76% within the four systems in experimental Phases I, II and III, respectively (Table 6). With the increase of inflow wastewater strength in experimental Phases II and III, relatively high mean removal of TN within the range of 48–76% was detected in all four systems. TN removal of only 55% was also observed using an aerated CW system in another study [71]. A relatively high TN in the outflow dynamics (Figure 4) and mean concentrations in the range of 89–190 mg/L were observed in experimental Phase III with high-strength wastewater with a mean inflow TN concentration of 374 ± 138 mg/L (Table 6). This was potentially due to the overestimation of TN (target TN was 240 mg/L, Table 1) and thereby overdose of the chemicals for increasing TN concentration in the inflow within this phase.

A relatively high NO₃-N mean concentration (46–101 mg/L) in the outflow of all four systems suggested high nitrification and prevailed aerobic conditions that potentially inhibited denitrification, and therefore a relatively low mean TN removal efficiency was observed in high-strength wastewater as inflow in experimental Phase III. Therefore, an effective TN removal did not occur, presumably due to a lack of available carbon source and anoxic conditions (e.g., lack of denitrification) within the four systems. Together with the knowledge that the presence of oxygen inhibits both the synthesis and activity of denitrification enzymes [72], this study’s results recommend the denitrification process within the artificially aerated systems as the minor nitrogen removal pathway. The removal of TN up to 96% was recorded in a laboratory-scale aerated constructed wetland through configurational differences where only the front end of the wetland bed was aerated, thus providing anoxic zones in the other end [70]. Thus, optimum NO₃-N or TN removal could be achieved by limiting the aeration to a fraction inside the systems. Intermittent aeration can potentially contribute to the DO fluctuations and can form both aerobic and anoxic regions inside the systems to promote higher TN removal. Nevertheless, limited TN removal in vertical flow wetlands (only 20–30%) is commonly acknowledged in the literature as well [73]. However, an unintentional overdose of nitrogen compounds in the modified wastewater inflow also played a role in a relatively low TN removal by the systems when operating with high-strength wastewater, and none of the outflow mean TN concentrations were in compliance with the Jordanian class A (<45 mg/L) or even class B standard (<70 mg/L) in experimental Phase III within this study (Table 4 and Table 6).

3.1.5. TP Removal

During the experimental phases from I to III with a mean inflow TP concentration of 7.9 ± 3, 18.8 ± 5 and 45.3 ± 12 mg/L, the mean TP removal efficiency in the range of 17–69% in the outflow of the four systems was observed (Table 6). The dynamics of inflow and outflow TP concentrations in the three experimental phases with different wastewater compositions are also shown in Figure 5.

With the increase of inflow wastewater strength, specifically operating with high-strength wastewater in experimental Phase III, relatively high removal of TP within the range of 48–69% was observed in the systems, except for the MBR with a mean TP removal efficiency of 18% only (Table 6). The rapid increase of TP concentration in the inflow also presumably resulted in a remarkable increasing trend of higher TP concentrations or fluctuations in the outflow dynamics of all four systems (Figure 5). Therefore, increasing the organic loading may potentially have led to improved phosphorus removal in this study. Except for the MBR system, the mean concentrations of TP in the effluent of the MBBR, SBR and AVFCW systems were significantly below the Omani standard of class A, which is 30 mg/L for reuse in agriculture (Table 4).

3.1.6. Removal of E. coli

The profiles of E. coli counted within the inflow and outflow samples in all three experimental phases are shown in Figure 6.

During operation in experimental Phases I, II and III with the mean inflow E. coli of 7.1 ± 0.3, 6.9 ± 0.4 and 6.2 ± 0.6 log10 (MPN/100 mL), respectively, decreased down to a mean outflow E. coli in a range of 0.1–4.8 log10 (MPN/100 mL). This accounted for a mean removal of E. coli in a range of 2.3–6.9-log removal from the four systems in this study (Table 6). During the experimental Phases II and III with a stepwise increase of wastewater strength, the E. coli count demonstrated a slight downward trend in the outflow dynamics of all four systems (Figure 6). As expected, the MBR system showed almost a complete removal of E. coli in a range of 6.1–6.9-log removal in the outflow during all three experimental phases and performed the best for pathogen removal as compared to the other three systems in this study. The other three systems (MBBR, SBR and AVFCW) achieved E. coli removal in the range of 2.8–4.8-log removal.

Apart from the MBR (<1 MPN/100 mL), in experimental Phase III with high-strength wastewater, the E. coli counts from the other three systems were not in accordance with the agricultural reuse standard of class A water quality as suggested by the Jordanian, Omani and European Union regulations. A mean outflow from the MBBR and AVFCW systems with 3.8 ± 0.6 and 3.6 ± 0.5 log10 E. coli, respectively, met the water quality standard of class D (<10,000 MPN/100 mL) for agricultural reuse suggested by the EU regulations (Table 4 and Table 6). The SBR system with 2.8 ± 0.8 log10 E. coli met the treated water quality class C (<1000 MPN/100 mL) for agricultural reuse according to the Jordanian, Omani and EU standards.

E. coli removal in the AVFCW was somehow limited, with an average reduction of 2.6-log removal. This is partly similar to E. coli removal reported by Headley et al. [74], with saturated vertical flow wetlands with aeration as 2.1-log removal. For higher pathogen removal, more optimization and final polishing would be necessary for the MBBR, SBR and AVFCW systems. A slow-sand filtration as a post-treatment step, particularly for the AVFCW system, or the inclusion of an ultraviolet (UV) lamp at the outlet would have been very effective for pathogen removal, depending on the required treated water quality class for reuse. The AVFCW can also be considered in combination with a horizontal flow CW for better E. coli and nitrogen removal.

It is important to highlight that, in this study, the E. coli counts in the modified high-strength wastewaters were reduced instead of getting higher in more concentrated wastewaters (Figure 6). Since the wastewater discharge rates were incrementally reduced (halved) in each phase, and the reduced volume was simulated by adding chemicals to increase the concentrations of specific parameters, the number of E. coli was reduced in more concentrated wastewater. In reality, it was difficult to artificially increase the number of E. coli in the influent of modified high-strength wastewater, and this is potentially an important limitation of this study that should not be overlooked.

3.2. Comparing the Effectiveness of the Selected DWWT Technologies

The reuse of treated wastewater in agricultural irrigation should ensure compliance with higher water quality classes or categories (class A, B, C or D) according to local or international standards. To achieve this, effective treatment technologies need to be selected, operated and practiced on various decentralized applications; reuse protocols or guidelines should also be standardized [75].

In general, the four selected small DWWT technologies used in this study performed efficiently when treating high-strength wastewater with extremely high organic and nutrient concentrations, which were significantly beyond the EU practical testing ranges. Results also showed very high NH₄-N removal efficiencies. However, the technologies did not remove NO₃-N, TN, TP and E.coli to a great extent when treating high-strength wastewater.

To compare the DWWT technologies, one important advantage of the MBR system is that it has the ability to be amazingly effective in removing pollutants and can be an ideal option for decentralized wastewater treatment and safe reuse applications [76]. The MBR systems provide excellent effluent quality with reduced footprint and lesser sludge generation as compared to other conventional DWWT technologies [77]. Despite that, their major disadvantage is associated with membrane fouling [23], which potentially increases maintenance and operating costs or membrane regeneration costs. The MBR system used in this study also demonstrated very good performance in treating high-strength wastewater and produced high-quality treated water with efficient removal of BOD₅, COD and TSS, and nearly complete removal of E. coli. However, the membrane fouling led to a complete clogging of the membrane, which was finally solved by replacing the clogged membrane with a new one. It clearly indicates that regular maintenance work needs to be performed to maintain a continuously high-performing MBR system. Otherwise, the membrane may lead to potential clogging due to high organic load, thus reducing treatment performance and requiring expensive membrane replacement for further water quality improvement. Consistent maintenance of the SBR and MBBR systems is also necessary for proper operation.

The AVFCW used in this investigation demonstrated a highly promising NBS with a very efficient treatment performance when treating high-strength wastewater. The presence of plants (Phragmites) and their intricate root structure may play a role in a higher mean DO within such intensified constructed wetland systems. The plants showed some stress when operating with high-strength wastewater in experimental Phase III, and their growth decreased gradually under extreme weather, for example, during hot summer days and during winter. However, the system proved to be very robust, requires simple operation and maintenance, and can be ideal for decentralized applications in small communities in rural areas. Further research should aim to optimize the system in order to increase TN and E. coli removal and to determine how aerated systems respond under intermittent aeration.

Treated water quality is an important factor for various reuse purposes (e.g., agricultural irrigation), and therefore assessing the local regulations and allowable limit values for the safe reuse of treated wastewater is very important. Such regulations and guidelines on small DWWT technologies would potentially recognize the importance of decentralized wastewater treatment for arid/semi-arid regions (e.g., Jordan or Oman) and would definitely be helpful in promoting decentralized wastewater treatment and reuse on a local scale [13]. A good compilation of a variety of existing national standards or guidelines for treated wastewater reuse can be found in [78].

This study’s results propose that small DWWT technologies can be implemented for wastewater treatment and agricultural reuse purposes in the remote areas of arid/semi-arid regions, which can provide an additional water resource for countries with high water scarcity such as Jordan, Oman, etc. However, long-term operation under local climatic conditions and further optimization of these DWWT technologies should be mandatory for more sustainability [31] and to meet local regulatory limit values for safe agricultural reuse of the treated effluent.

4. Conclusions and Recommendations

This study has clearly shown that the EU-certified small DWWT technologies can efficiently treat high-strength wastewater with concentrations beyond the EU-testing ranges (i.e., BOD₅ > 500 mg/L; COD > 1000 mg/L; TSS > 700 mg/L) and can be promising wastewater treatment alternatives for small communities in the rural areas of arid/semi-arid regions. During the three experimental phases with different wastewater compositions, all four selected systems showed highly efficient removal of BOD₅, COD, TSS and NH₄-N, but showed a rather limited performance in the case of NO₃-N, TN, TP and E. coli (except MBR) removal. However, comparatively more NO₃-N and TN were removed in the MBBR system, while more E. coli were removed in the MBR system. The final outflow mean BOD₅, COD and TSS concentrations were in compliance with class A standards as outlined in the Jordanian and Omani national standards for agricultural reuse. Maximum allowable limit values for Class A reuse standard were not achieved for other recommended parameters by the systems but met other quality classes or categories (e.g., class B, C and D). In general, the selected DWWT technologies can efficiently remove organic pollutants from high-strength wastewater and reduce the substantial risk of pathogens; and the treated effluent can be reused for both restricted and unrestricted irrigation purposes (depending on the water quality requirements). The stepwise increase of inflow wastewater strengths showed neither any remarkable impact on the overall treatment performance of the DWWT technologies nor a limiting factor for stable operation if sufficiently maintained. However, long-term experimental operation and further optimization as well as potential post-treatment steps should be necessary. Further research should aim to simultaneously meet the highest quality of all reuse standard parameters, including E. coli, which could potentially be achieved with a subsequent treatment step such as a sand filter or the use of an ultraviolet (UV) disinfection unit.

Treating wastewater using small DWWT technologies in combination with the safe reuse of the treated effluents for agricultural irrigational purposes can potentially create noticeable environmental and economic benefits in arid and semi-arid regions. These DWWT technologies are not only designed to operate at small scale for a smaller community or household basis to reduce the effects of wastewater disposal on the nearby environment and to protect public health but also increase the potential for reuse the treated wastewater in agriculture and food production, depending on the local community, reuse regulations, appropriate system selection and technical feasibility. Therefore, this study fosters a broader perspective on the treatment performance of small DWWT technologies and helps to overcome the difficulties that the decision makers and administrators face in assessing, permitting and implementing such systems in a given local context.

A localized certification guideline for the EU-certified DWWT technologies based on local needs (e.g., in Jordan and Oman) and socio-economic and climatic conditions should be mandatory. Guidelines to ensure minimum quality and performance standards of DWWT technologies for wastewater treatment and reuse, requirements for O&M of such systems, requirements for local training and administrative protocol should be developed before the final permitting process. Applying a methodology for the up-scaling of such small DWWT technologies to ensure conformity between the pilot test and full-scale technology is also recommended.

Overall, this study demonstrates the possibilities of using small DWWT technologies to transform high-strength wastewater from a major environmental and health hazard into a potentially clean and attractive resource.