Utilization of arti ﬁ cial recharged ef ﬂ uent for irrigation: pollutants ’ removal and risk assessment

The reclaimed water from soil aquifer treatment (SAT) column was reused for irrigation as the source water, pollutants ’ removal and health risk assessment was analyzed via the comparison with secondary and tertiary ef ﬂ uents. The effect of the SAT pre-treatment on the qualities and growth of different crops ( Lachca sativa – lettuce, Brasica rapa var chinensis – pak choi, Cucumis sativus – cucumber, Brassica oleracea – cabbage, and Zea mays – maize) were evaluated. Experimental results demonstrated that the tertiary and SAT treatments had no signi ﬁ cant effect on the crop qualities, and could ef ﬁ ciently decrease the accumulation of heavy metals (especially for SAT pre-treatment). Moreover, the carcinogenic risk of the chemical carcinogens for the 1.5 m SAT ef ﬂ uent irrigation declined roughly an order of magnitude as compared with the secondary ef ﬂ uent, and three to four orders of magnitude decreasing of the virus risk. These ﬁ ndings are signi ﬁ cant for the safe and cheap reuse of secondary ef ﬂ uent for irrigation purposes.


INTRODUCTION
Water shortage is a serious environmental issue in arid and semi-arid regions that requires the exploration of water reuse options (Bakopoulou et al. ; Wei et al. ). To solve the water crisis throughout the world, rainfall, deep groundwater, seawater, etc., have been recognized as new alternative water resources (Elimelech & Phillip ; Erban et al. ; Jung et al. ). However, exploitation of the above water resources is always restricted by the local, economic and technological conditions, which limits their further application (Busch & Mickols ; Henriques et al. ). Considering that municipal wastewater treatment plants (WWTP) produce stable and abundant flows during different seasons, development of secondary effluent as an attractive alternative for water reuse is practically urgent (Kalkan et al. ; Zucker et al. ). Thus, how to efficiently recharge and then reuse the secondary effluent is rapidly becoming a necessity for many municipalities throughout the world (Wei et al. ). Generally, the traditional reuse approaches of secondary effluent include mainly agricultural irrigation, industrial processing, cooling water, toilet flushing, wetland habitat creation, restoration and maintenance, groundwater recharging, landscape water replenishment, among others (Bunani et al. ).
China is one of the largest agricultural countries in the world and the amount of water used for agriculture irrigation accounts for about 70% of the total water consumption (Li et al. ; Zhao et al. ); However, the lack of water sources constrains the development of agriculture (especially for northwest China), and the increase in population and shrinking of water supplies has strengthened this water shortage (Huang et al. ). Recently, the search for a renewable water source, such as secondary effluent, for agricultural irrigation is considered to be practically applicable. However, a relatively high level of heavy metals, persistent organic pollutants, salt ions, and other elements, as well as a certain amount of pathogenic bacteria have provided limits to its direct agricultural reuse (Rizzo et al. ).
Earlier work by Lado et al. () demonstrated that a significant accumulation of total organic matter, Cu, Ni, Zn and B in the top 1.0-1.5 m soil layer during long-term irrigation (>7 yr) with secondary treated wastewater, might be potentially toxic to some sensitive crops. Friedman et al. () observed that higher levels of nitrogen and Mn were accumulated in celosia and phosphorus in sunflower during irrigation with secondary-treated municipal effluents.
Recent work by Bakopoulou et al. () pointed out that secondary effluents produced in the Thessaly region are suitable for irrigation reuse, especially for crops which are not used raw by humans; however, the seasonal and sitespecific inhibition of growth on the plants were noted. Oron et al. () stated that a UF/RO system would significantly improve the water quality of the secondary effluent, and minimize the health risk of agricultural products which had been irrigated with reclaimed wastewater. Generally, the existence of heavy metals, salinity, sodium, residual chlorine and other containments could partially affect plant growth, and consequently negatively affect the crop characteristics and soil properties. To effectively remove these hazardous pollutants in the secondary effluent, pre-treatments such as coagulation, ozone oxidation, granular activated carbon adsorption and membrane filtration have been applied as tertiary treatments (Bixio & Wintgens ; Kalkan et al. ; Pramanik et al. ). Generally, the above-mentioned pre-treatments are expensive and need chemical additions or regeneration steps. Therefore, less expensive and environmentally friendly methods for secondary effluent pre-treatment are desired.
Soil aquifer treatment (SAT) is an artificial water recharge technique frequently used to renovate domestic effluents for potable and non-potable purposes (Xue et al. ; Schaffer et al. ). The water quality is substantially improved via the integrated functions of biological, chemical, and physical processes in the aquifer (Hübner et al. ). To this end, the SAT system may be an alternative approach to the treatment of secondary effluent, and it has the advantages of lower cost, easier operation and greater efficiency. The SAT system can store the reclaimed water in the aquifer layer, with no evaporation losses and also can achieve inter-seasonal and inter-year storage (Dillon et al. ). Although several studies in the past years have attempted to explore secondary effluent as a water resource for agriculture irrigation, to the authors' knowledge seldom do researchers mention the application of SAT as a pretreatment for secondary effluent irrigation.
The goal of this work was to: (1) study the effect of SAT pre-treatment on water quality for irrigation; (2) evaluate the effect of SAT pre-treatment on production of typical agricultural crops; and (3) assess the possible health risk of reusing the reclaimed water for irrigation.

SAT system set-up and operation
Laboratory-scale soil column system which simulated aquifer conditions in a series of three 55 cm columns (diameter 10 cm) was constructed and operated for a period of more than two years, which was operated with a cycle of 16 h wetting/8 h drying. The influent was pumped upwards through the column at a desired flow rate of 15 mL/h controlled by a peristaltic pump. A complete description of the operation parameters of the SAT system is given in Xue et al. ().
Experimental soil samples were collected from the dry bed of Songhua River in Harbin (potential recharge sites). After air drying, soil samples with particles greater than 2 mm were sieved out, then packed into the columns and further compacted to field density. The soil samples had an average pH of 8.2, organic carbon content (OC) of 2.9%, cation exchange capacity of 7.4 cmol/kg, and soil composition of 49.3% sand, 44.5% silt, and 6.2% clay, respectively.

Simulation of agricultural irrigation of crops
Secondary, tertiary (coagulated-filtrated) and the SAT efflu-

Accumulation of heavy metals in the crops after irrigation
The crops cucumber, cabbage, and maize used for the irrigation study were planted in spring and harvested in autumn, and irrigated in the same condition (irrigation was performed once 3-5 d, 2 L water/strain). For comparison, the secondary effluent, tertiary effluent, and SAT effluent (0.5, 1.0, and 1.5 m) were selected as the irrigation water samples.
Once the crops were harvested, each of the samples was immediately collected and homogeneously mixed and then stored in plastic bags for further analysis.

Chemical analysis
All the collected water samples were filtrated using 0.45 μm cellulose nitrate membrane filter and stored at 4 W C prior to analysis. Parameters, namely, pH, water content, organic matter, COD, soluble COD (SCOD), total nitrogen (TN),

RESULTS AND DISCUSSION
Pollutants' removal during the artificial SAT system operation As shown in Table 1, more than 70% of NH þ 4 -N, TP, biochemical oxygen demand (BOD 5 ), Cr, Cd, As, Zn, Ni, and   In comparison to the tap water, the irrigation with secondary, tertiary, and SAT effluents increased the productivity of lettuce and pak choi slightly. For example, the average leaf area of the pak choi was 10.7 cm 2 for secondary effluent irrigated samples and 9.4 cm 2 for SAT effluent; this is much higher than that of the tap water irrigated samples (7.8 cm 2 ). As shown in Figure 2, the wet weight of lettuce averaged 0.51 g after 40 days' irrigation with secondary effluent, 0.43 g for tertiary effluent, and 0.44 g for SAT effluent, respectively, with that of the tap water being the lowest where P i is the lifetime risk caused by chemical non-carci-
The health risk of the different effluent irrigations via the food chain can be calculated based on the hypothesis that the exposure frequency is once per day, with an exposure dosage of 10 mL. As shown in Table 2, the calculated annual risk of the chemical non-carcinogens during the secondary effluent irrigation was 1.18 × 10 À11 , and decreased to 7.62 × 10 À12 for tertiary effluent, and further declined to 3.64 × 10 À12 for the 1.5 m depth SAT effluent. It is obvious that irrigation using the SAT effluent would significantly decrease the health risk of chemical non-carcinogens to farmers, and is meaningful for agricultural irrigation purposes.
As shown in Table 3, the carcinogenic risk caused by inhalation during the secondary, tertiary, 0.5 m depth SAT and 1.5 m depth SAT effluents' irrigations were 5.69 × 10 À7 , 4.31 × 10 À7 , 3.50 × 10 À7 , and 4.58 × 10 À8 , respectively, and are much lower than the threshold value (5.0 × 10 À5 /yr) recommended by ICRP. Overall, the carcinogenic risk of the chemical carcinogens during the SAT effluent irrigation was approximately an order of magnitude lower than the irrigation with secondary effluent. Moreover, element Cr contributed as much as 80% of the bulk carcinogenic risk during the SAT effluent irrigation, and was the predominant precursor of the chemical carcinogens within the SAT effluent.

Virus risk assessment during irrigation with different effluents
The pathogenic bacteria found in the irrigated water samples were mainly intestinal virus (e.g., Coxsackie, rotavirus and hepatitis A virus), and the probability of infection of these virus can be calculated based on the beta-Poisson model as: Assuming that the annual exposure frequency of the irrigated water to a farmer was 40 times per year, with an exposure dosage of 1 mL every time, the corresponding infection probability of the secondary, tertiary, 0.5 m depth SAT and 1.5 m depth SAT effluents were calculated and the results are listed in Table 4.
As shown in Table 4, the annual risk of rotavirus to the farmer was 6.72 × 10 À4 for the secondary effluent irrigation, which exceeded the maximum acceptable value of EPA (10 À4 /yr) for individual risks. For comparison, the calculated annual risk of the tertiary effluent irrigation Hepatitis A virus 5.787 × 10 À6 2.314 × 10 À4 4.547 × 10 À8 1.819 × 10 À6 1.535 × 10 À8 1.075 × 10 À6 9.211 × 10 À10 6.448 × 10 À8 Coxsackie 6.652 × 10 À8 2.660 × 10 À6 4.410 × 10 À10 1.764 × 10 À8 1.488 × 10 À10 1.042 × 10 À8 8.929 × 10 À12 6.250 × 10 À10 declined to 5.28 × 10 À6 , and further to 1.83 × 10 À7 for the 1.5 m SAT effluent. Similarly, the annual risk of the hepatitis A virus to farmers during the secondary effluent irrigation also exceeded the EPA benchmark, indicating that the typical virus in the secondary effluent should be of a high concern for its potential human health risk. As expected, the health risk of hepatitis A virus during the tertiary and SAT effluents' irrigation was much lower, and both can meet the requirement of the criteria of EPA. Generally, the annual risk of a typical virus during the SAT effluent irrigation was three to four orders of magnitude lower than that of the secondary effluent irrigation, clearly demonstrating that the control of health risks during the secondary effluent irrigation would benefit from the pretreatment of SAT.

CONCLUSION
The following conclusions are drawn based on the presented experimental results: 1. SAT system exhibited a relative higher removal efficiency of most pollutants within the secondary effluent as compared to the tertiary treatment, especially for the reduction of NH 4 þ -N, TP, BOD 5 , Cr, Cd, As, Zn, Ni, and Mn.
2. Irrigation with secondary, tertiary, and SAT effluents enhanced the growth of pak choi, with a root growth inhibition rate of À5.9% for secondary, À21.6% for tertiary, and À23.2% for SAT effluents, respectively. The noticeable toxicity of secondary effluent was observed for the growth of lettuce with a positive root growth inhibition rate of 21.3%, while the irrigation with the tertiary and SAT effluents enhanced the growth of lettuce. Moreover, irrigation with secondary, tertiary, and SAT effluents led to slight increases in the productivity both for lettuce and pak choi.