*Article* **Specific Types and Adaptability Evaluation of Managed Aquifer Recharge for Irrigation in the North China Plain**

#### **Shuai Liu 1, Weiping Wang 1,\*, Shisong Qu 1, Yan Zheng <sup>2</sup> and Wenliang Li <sup>1</sup>**


Received: 30 September 2019; Accepted: 14 February 2020; Published: 18 February 2020

**Abstract:** The North China Plain is the main grain production district in China, with a large area of well irrigation resulting in a large groundwater depression cone. In the 1970s and 1980s, small-scale managed aquifer recharge (MAR) projects were developed to recharge shallow groundwater, which played an important role in ensuring stable and high crop yields. MAR projects are divided into 10 types based on local water conservancy characteristics. The combined use of well–canal irrigation has been widespread in the Yellow River Irrigation District of Shandong Province for nearly 40 years, where canals play multiple roles of transporting and storing Yellow River water or local surface water, recharging groundwater and providing canal irrigation. Moreover, the newly developed open channel–underground perforated pipe–shaft–water saving irrigation system can further expand the scope and amount of groundwater recharge and prevent system clogging through three measures. Finally, an adaptability zoning evaluation system of water spreading has been established in Liaocheng City of Shandong Province based on the following five factors: groundwater depth, thickness of fine sand, specific yield, irrigation return flow, and groundwater extraction intensity. The results show that MAR is more adaptable to the western region than to the eastern and central regions.

**Keywords:** types ofMAR for irrigation; Yellow River Irrigation District; adaptability zoning evaluation

#### **1. Introduction**

Managed aquifer recharge (MAR) is the intentional recharge of water to aquifers for subsequent recovery or environmental benefits, and MAR projects must achieve effective aquifer recharge under different terrains, hydrogeological conditions, water sources, and water demand characteristics [1]. As an effective water resources management measure, MAR has been widely used in many countries, especially in semi-arid and arid areas [2–4]. The North China Plain is the main grain-producing area in China. The average annual precipitation ranges from 500 to 900 mm, and the majority of the annual precipitation occurs from July to August. Agricultural irrigation requires a large amount of water and mainly relies on groundwater extraction, resulting in a large area of groundwater overexploitation. In the 1970s and 1980s, a variety of small-scale MAR projects were developed to recharge shallow groundwater, and they played an important role in ensuring stable and high yields of crops [5]. However, these small-scale MAR projects lack scientific review and evaluation.

In this study, 10 types of specific agricultural MAR in the North China Plain are summarized. Two methods have been developed in the Yellow River Irrigation District to apply and develop water spreading methods: a well–canal combination mode and an open channel-underground perforated pipe-shaft-water saving irrigation system. These two modes are both effective for sustaining a high

grain yield and restoring groundwater overexploitation. Based on the special hydrogeological features of phreatic water (0–60 m) in the Yellow River Irrigation District of Shandong Province, a main aquifer of fine sand is determined to be suitable for water spreading instead of recharge through a well. However, the application of water spreading is influenced by many factors. Which area is suitable for it? Liaocheng City was selected to establish an adaptability zoning evaluation system for the MAR of water spreading, and five factors were evaluated: groundwater depth, thickness of fine sand, specific yield, irrigation return flow, and groundwater extraction intensity. Through the combination of GIS and the DRASTIC model, this adaptability zoning evaluation provides a scientific basis for the sustainable development of the MAR projects in the Yellow River Irrigation District [6–8].

#### **2. Progress of MAR**

#### *2.1. MAR Application of Agriculture in the North China Plain*

China has made many achievements in aquifer recharging, especially in industrial and agricultural production, urban water supply, and ecological protection [9–11]. The characteristics of MAR as a component of irrigation and drainage systems are as follows: small scale, short service life, low investment, fixed irrigation water sources, and crop planting structures. Considering these characteristics and the relationship between surface water and shallow aquifers, MAR is divided into three types: water spreading, well recharging, and a combination of both. MAR can be further divided into 10 types according to the specific farmland water conservancy project (Figure 1).

#### 2.1.1. Water Spreading

The term "water spreading" refers to the release of water over the ground surface to increase the quantity of water infiltrating into the ground and percolating to the water table. Its characteristics are large influence areas, low investment, and high efficiency. It is divided into three types: field infiltration, infiltration pond, and infiltration ditch.

(a) Field infiltration: Field infiltration is achieved by check irrigation, flood irrigation, and large-scale winter irrigation, and it has the characteristics of steady water distribution and a wide range of influence. The amount of recharge is related to the irrigation quota and groundwater depth. <sup>1</sup> Check irrigation: Water is diverted into fields with borders, and the thickness of the water layer is approximately 0.33 m. This method is carried out in winter wheat fields and white stubble land. Especially in coastal saline–alkali areas this method can be used to wash out soil salinity, dilute groundwater quality, and replenish groundwater. <sup>2</sup> Flood irrigation: After water is injected into the recharge area through ditches, it is controlled by earthen embankments to make the water overflow. This method is suitable for flood plains with low slopes, old channels, sandy wastelands, woodlands, and orchards. <sup>3</sup> Winter irrigation with a large irrigation quota: In areas without check irrigation or flood irrigation conditions, large-quota winter irrigation can be used to recharge groundwater.

(b) Infiltration pond: Surface water is allowed to go through the unsaturated zone into the aquifer through natural ponds that are renovated and connected to conveyance canals. The characteristics of these ponds are low land occupation, a large amount of recharging, and a limited range of influence.

(c) Infiltration ditch: Infiltration ditches are the major method of recharging aquifers in North China. This method requires the selection of a reasonable ditch spacing. The infiltration water first forms a water peak along the ditch, then spreads from the ditch to both sides before reaching the center of the ditch.

(d) Ditch–underground permeable cement pipe-pond system: This system can be used in places with low permeability and a shortage of land. The diameter of the underground permeable cement pipe is more than 30 cm. A pond for desilting and lifting irrigation is built at a location 100 m away from the pipe head. When the pipe is buried as shallow as possible, it can also infiltrate the tilth topsoil and play an important role in irrigation.

(e) Tunnel–well: This system consists of four parts: river, tunnels, wells, and artificial water lifting facilities. A tunnel with a width of 0.8 m and a height of 1 m is connected to the river and located in the clay layer 7–8 m below the ground. Each tunnel is several kilometers long, and there is a well for water lifting every 30 m. During the flood season this system can increase the amount of recharging and raise the groundwater table rapidly, with a better infiltration effect than those of other facilities.

#### 2.1.2. Well Recharging

(f) Seepage well: Water can flow directly into the aquifer through shallow wells. This system, which is suitable for areas with deeply buried gravel aquifers, is faster than water spreading and has low land occupation, although clogging frequently occurs.

(g) Shaft well: The vertical shaft is used to expose the soil layer so that the water is admitted directly into the sandy gravel aquifer with high permeability. However, the bottom silt needs to be cleared regularly. When combined with (b) infiltration pond, better efficiency is observed.

**Figure 1.** *Cont.*

**Figure 1.** Specific types of MAR in agricultural irrigation.

#### 2.1.3. Combination Methods

(h) Canal–pipe–well: The pumping well, which is connected to the canal by a pipe, lifts groundwater when the canal is waterless. The slope of the canal is gentle in the design, and the intake is more than 0.5 m above the bottom of the canal to reduce sedimentation.

(i) Brackish aquifer treatment: Brackish water is pumped through wells, and fresh water is diverted, stored, and infiltrated by deep ditches. The ditches can also be used for irrigation. Soil salinization can be controlled, and brackish aquifer can be desalinated.

(j) Ditch–well–check gate system: In plain areas, a number of check gates are added to ditches based on border checking, reticulated canal systems, and driven well groups. This system represents a farmland water conservancy system of diversion, storage, infiltration, water-saving, irrigation, and drainage.

#### *2.2. Cases*

#### 2.2.1. Well-Canal Combination in the Yellow River Irrigation District

Overview of the Yellow River Irrigation District in Shandong Province

The Yellow River has a drainage area of 795,000 km2 and a length of 5464 km, making it the second-longest river in China. The Lower Yellow River flows through six cities in Shandong Province and ends at the Bohai Sea in Lijin County, covering about one-third of Shandong Province [12]. The area has a warm temperate monsoon climate. The average annual rainfall is 606 mm, and the average annual evaporation is 1300 mm. Rainfall is mainly concentrated in the summer; thus, the area presents characteristics of spring and autumn droughts and summer floods. The Yellow River flood plain is the main part of the North China Plain. Due to the long-term geological movements and the sedimentary rhythm in the downstream area of the Yellow River, the formation is deep and the aquifer media are dominated by fine sand with a hydraulic conductivity of 0.3–2.5 m/day [13]. The area can be divided into three aquifers from top to bottom. The most important water supply for agriculture is phreatic water, with a buried depth of 0–60 m, a roof depth of 10–25 m, a thickness of 10–25 m, and a well yield of 6–10 m3/h·m. The lithology of the unsaturated zone and the characteristics of the shallow aquifer provide favorable conditions for recharging aquifers and exploiting shallow groundwater. The Yellow River Irrigation District of Shandong province is shown in Figure 2. The designed irrigation area of the Yellow River of Shandong province is 16,667 km2. The average annual water diversion for agriculture in the Yellow River Irrigation District is 4.5 billion m<sup>3</sup> (1995–2015). The main crops are winter wheat, corn, and cotton, with a multiple cropping index of 1.7. With the influence of inadequate local water resources, unreliable Yellow River water available in the low Yellow River flow years, and uneven distribution of water between upstream and downstream of the Yellow River Irrigation District, the shallow groundwater overexploitation area has reached 4500 km2. In short, the Yellow River Irrigation District is facing drought, flooding, and sediment problems [14].

**Figure 2.** Yellow River Irrigation District of Shandong Province.

The Yellow River Irrigation District of Shandong Province is 16,667 km2, and only 12% of this area uses gravity irrigation. Therefore, three irrigation-drainage modes have been established. (1) Irrigation and drainage systems are separated and use gravity flow. The canals of the irrigation system (including the main, branch, lateral, and sublateral canals) are above ground, while the corresponding ditches of the drainage system are underground. The drainage water enters the Bohai Sea through the river. (2) Irrigation and drainage systems are separated, and water is pumped from the canal for irrigation and gravity drainage. The canals of the irrigation system and the ditches of the drainage system are underground. The ditches are deeper than the canals to ensure that the drainage water can flow into the river under the influence of gravity. (3) The well–canal combination, which is the most commonly used mode of irrigation and drainage, is a combination of irrigation and drainage systems.

#### Well-Canal Combination Mode

With the urbanization and industrialization of the Yellow River Irrigation District, the water distribution has shifted from agricultural irrigation to urban water supplies. Yellow River diversion occurs over a limited time; thus, although the Yellow River water is available, the flow is unreliable. Therefore, the well–canal combination mode is inevitably adopted to achieve a large-scale steady production increase in agriculture [15–17]. The well-canal combination mode is a double irrigation system, which has the advantages of recharging aquifers with river water through canal infiltration and irrigation return flow, and guaranteeing bumper harvests with wells (Figure 3). The canal has multiple functions of water delivery, storage, infiltration, irrigation by pumping, and drainage [18].

**Figure 3.** Well–canal combination mode (**a**) with Yellow River water diversion and (**b**) without Yellow River water diversion.

Historically, flood irrigation was widely used in the fields of the Yellow River Irrigation District; therefore, irrigation return flows became the main source of groundwater recharging. The irrigation efficiency and overall benefit of the Yellow River Irrigation District are low. At present, low-pressure pipeline irrigation has been widely used, resulting in a lower irrigation quota and irrigation return flow, a higher irrigation application efficiency, and lower groundwater pollution by soil leaching. The amount of groundwater recharged in this way is much less than that from flood irrigation. The method mainly depends on the leakage from canals after implementing the water saving measures, and does not pass through the plow layer and thus does not pollute the shallow groundwater. The wide use of the well–canal combination maintains the balance of diversion water and groundwater and has a good effect in regulating the groundwater table.

Figure 4 shows the water balance of the Yellow River Irrigation District with a 50% probability of precipitation. When the probability of precipitation is 50%, the total water demand of the main crops is 900 mm and the precipitation is 606 mm, of which 182 mm is infiltrated into the ground. Therefore, the water deficiency is 294 mm. The 269 mm of Yellow River water diversion can fill the water deficiency. The irrigation efficiency of Yellow River Irrigation District is 0.6. The average annual available local water can only guarantee 67% of the irrigation water requirement, and the rest needs to be provided by the Yellow River, of which the field irrigation volume is 161 mm and accounts for 60% of the total amount of Yellow River Diversion, and the remaining 108 mm accounts for 40% of the total, which can be pumped by well to irrigate crops for harvesting from the groundwater recharged through canal system infiltration and irrigation return flow. The Yellow River water diversion has played a key role in stable agricultural production and high yields. When the irrigation water ratio of well to canal is 1:0.93, the irrigation water in this area basically maintains a balance of supply and demand. In addition, other research has shown that when the irrigation water ratio of well to canal is at 1:0.78 in

the People's Victory Canal Irrigation District of neighboring Henan Province, it can basically maintain the balance of supply and demand [19]. If water balance is reached in the water deficient area by Yellow River water diversion, this mode is influenced by irrigation efficiency. The irrigation water ratio of well to canal would decrease with increases in irrigation efficiency, resulting in a decreased capacity for groundwater regulation. So, the question is how to enlarge the groundwater recharge amount through Yellow River water diversion over a limited time [20–22].

**Figure 4.** Water balance of the Yellow River Irrigation District at 50% probability of precipitation.

#### 2.2.2. Open Channel-Underground Perforated Pipe-Shaft–Water Saving Irrigation System

With the large-scale promotion of water saving practices in the Yellow River Irrigation District, the flood irrigation method is no longer suitable. However, the canal system has already been formed. If new canals are dug, then they will occupy a large amount of cultivated land. Therefore, considering the existing irrigation-drainage system, a method of increasing the recharge rate of shallow groundwater overexploitation areas must be identified.

An open channel–underground perforated pipe–shaft–water saving irrigation system with high efficiency and ecology has been developed based on infiltration ditches, the combined tunnel–well mode, modern materials, and filtration technology (Figure 5). The underground perforated pipe is used to further increase the recharge rate and influence area, and to restore the groundwater overexploitation zone. The system consists of the Yellow River water source, open channels, prefilters, perforated pipes, dredging shafts, crops, irrigation systems, ambient groundwater systems, operation and monitoring facilities, equipment (e.g., electromagnetic flowmeters), and so on.

Three measures to prevent sediment blockage are set up in the recharging system. (1) A prefilter tank is provided at the canal head. (2) A 20-cm sand layer is placed around the pipe in the soil of the designed section, where the geotextile can prevent the outer sediments from entering into the pipe. The pipe has a certain slope and a dual function of water washing and seepage. (3) Shafts with well filters in the borehole are at the end of the recharge system, and they can not only receive the sediments but also contribute to the water seepage. The specific parameters of the recharge system are as follows: the bottom area of the prefilter tank is 4 m2, and the underground perforated pipe is composed of plastic blind ditches with a diameter of 30 cm, a slope of 1/500, a pipe length of 200 m, and a shaft depth of 10 m. The parameters can be adjusted according to different regions and local conditions.

**Figure 5.** Open channel–underground perforated pipe–shaft–water saving irrigation system.

In the practical project, the system was conducted at a recharge area of 54,000 m2 including three underground perforated pipes as subsystem and buried pipes spaced 90 m apart with 200 m lengths. According to the calculation by an analytical method, the project contributed a groundwater recharge amount of 52,310 m3, covering 60.7% of the total replenished water. The numerical simulation results showed that the groundwater table under the pipe rose 8.5 m throughout the simulation period. In the seven-day simulation period, the total infiltration amount of the open channel was 33,861 m3, and the total infiltration amount of the pipes was 56,224 m3, which accounted for 62.4% of the total infiltration amount [23]. In summary, this project has the advantages of no energy consumption, no land occupation, managed groundwater recharging, irrigation drainage, and water logging control, and it plays an important role in the recovery of groundwater funneling.

#### **3. Adaptability Zoning Evaluation**

#### *3.1. Study Area*

Liaocheng City is part of the North China Plain and is located in the northwestern part of Shandong Province and the western part of the Yellow River Basin, with good water diversion conditions (Figure 6). The terrain is flat, and the total area is 8715 km2. Liaocheng City is a large agricultural city with an average annual precipitation of 560 mm, a cultivated land area of 6353 km2, and an effective irrigation area of 4947 km2. The groundwater depth of shallow aquifers as irrigation water sources is 0–60 m. Most rivers are approximately parallel and flow from the southwest to the northeast. The rivers have a crossing and repeated sedimentary structure in the vertical direction. In the horizontal direction, the land can be divided into an accumulation area, a flood plain alternating area, and an interstream area, wherein the aquifer media become finer, the thickness decreases, and the water abundance weakens. The aquifer is composed of fine sand with a thickness of 10–25 m, fine sand with a thickness of 5–10 m, and silt with a thickness of less than 5 m. The main irrigation water sources are groundwater, water diverted from the Yellow River, and a small amount of surface water. Over the past 60 years, the number of agricultural irrigation wells has increased rapidly to guarantee bumper harvests. Groundwater continues to be overexploited, and the groundwater table is declining in part of this area. Therefore, to ensure that the combined use of surface water and groundwater sustain crop yields and also groundwater storage levels, the types and densities of recharge mechanisms need to

be adapted to the local intensity of water use and hydrogeological conditions. For these, zones were identified where water-spreading mechanisms were considered appropriate.

**Figure 6.** Location of Liaocheng City.

The constraint conditions for an adaptability zoning evaluation of MAR are as follows: (1) phreatic aquifers; (2) agricultural irrigation; (3) a Yellow River water recharge water source and a small amount of surface runoff during the flood season; and (4) the MAR project aim of a water spreading method, including indirect recharging methods such as field infiltration, infiltration ponds, ditches, and so on.

#### *3.2. Evaluation Factors*

An adaptability zoning evaluation means selecting a suitable recharge area to water spreading for irrigation, a process that is influenced by many factors. Based on the hydrogeological conditions of groundwater extraction, the evaluation focuses on the water-bearing characteristics of the vertical vadose zone during recharge. There are five factors for consideration: groundwater depth, thickness of the fine sand, specific yield, irrigation return flow, and groundwater extraction intensity.

Evaluation of groundwater vulnerability and recharge adaptability have similarities and differences in surface water infiltration. The process of the DRASTIC model can be used as a reference. The rating range for each evaluation factor is 1–5 points, for which the higher points correspond to a better adaptability of MAR for water spreading (Table 1). The overall score is calculated by equal weight.


**Table 1.** Grading of evaluating factors.

Because the five factors have different effects on groundwater recharge, they should be classified. The factors affecting the water storage capacity are groundwater depth, thickness of the fine sand, specific yield, and irrigation return flow (after decades of field experiments, abundant county-level data are available). The groundwater extraction intensity is equivalent to the user demand for recharge water. The lithology of the aquifer (0–60 m) is mainly fine sand, and the permeability coefficient is similar to that for a typical sand material. Therefore, the thickness of the fine sand can effectively represent the basic characteristics of Liaocheng City. Most data can be obtained from geologic reports and relevant departments, which are frequently used in groundwater resource assessments in the study area.

#### 3.2.1. Groundwater Depth

Groundwater depth refers to the status of groundwater overexploitation. Statistical data for 2016 indicated that there was an area of 898 km<sup>2</sup> with a groundwater depth greater than 18 m, and this accounted for 10.3% of the total area. This area has a large demand for recharge, and the highest evaluation score of 5 points was observed when the aquifer had a sufficient recharge time and amount of water. The area with a groundwater depth between 12 and 18 m was 1043 km2, which accounted for 12% of the total area, and the evaluation score was 4 points. The area with a groundwater depth between 6 and 12 m was 1240 km2, which accounted for 14.2% of the total area, and the evaluation score was 3 points. The area with a groundwater depth between 2 and 6 m was 5517 km2, which accounted for 63.3% of the total area, and the evaluation score was 2 points. The area with a groundwater depth less than 2 m was 16 km2, which accounted for 0.2% of the total area. Such areas have a small demand for recharge, and the score of this area was 1 point (Figure 7).

**Figure 7.** Corresponding areas of different groundwater depths.

#### 3.2.2. Thickness of Fine Sand

Thickness of fine sand refers to the water abundance characteristics of shallow aquifers. Values are the cumulative thickness of fine sand (0–60 m). The existing data were obtained by a large number of geological surveys conducted by the relevant departments. The area with a thickness of fine sand less than 5 m accounted for 15.1% of the total area. The water storage space was not large, and the score is the smallest at one point. The areas with thicknesses of 5–10 m and 10–15 m accounted for 11.4% and 17.3% of the total area and presented scores of 2 and 3 points, respectively. Areas with thicknesses of 15–20 m and greater than 20 m accounted for 38.4% and 17.7% of the total area and presented scores of 4 and 5 points, respectively.

#### 3.2.3. Specific Yield

Specific yield (μ) refers to the storage property of the formation (0–60m). A larger value of μ corresponds to a better water storage capacity. The corresponding values from 0.05 to 0.11 are based on the research results of the relevant departments. The specific yield can be obtained by the equation

$$
\mathfrak{u} = \frac{\mathfrak{a} \times \mathbb{P}}{\Delta \mathrm{h}}
$$

where α is the recharge coefficient of precipitation, P is the precipitation (mm), and Δh is the change in water level (mm).

#### 3.2.4. Irrigation Return Flow

Because the irrigation water quota is larger than individual rainfall under the same planting structure, groundwater depth, and uniform water distribution on farmland, irrigation return flow can better represent the hydraulic conductivity of unsaturated zones as the water spreading of MAR than with the recharge coefficient of precipitation.

$$
\beta = \frac{\Delta \mathbf{h} \times \boldsymbol{\pi}}{Q}
$$

where β is the irrigation return flow, Q is the amount of irrigation water (mm), Δh is the change in water level (mm), and μ is the specific yield of the area.

According to the above method, the relevant departments obtained the irrigation return flow through multiple irrigation experiments with an empirical value of 0.01-0.35. The larger the β is, the more permeable the unsaturated zone is.

#### 3.2.5. Groundwater Extraction Intensity

Groundwater extraction intensity not only refers to water demand but also to impact on a groundwater system. The empirical values of groundwater extraction intensity for irrigation in the study area were 9.1~18.3 <sup>×</sup> 104m3/a·km2. Groundwater extraction intensity was equal to the ratio of extraction volume to irrigation area.

#### *3.3. Results*

The maps of the five factors are as follows: (a) groundwater depth, (b) thickness of fine sand, (c) specific yield, (d) irrigation return flow, and (e) groundwater extraction intensity (Figure 8). The map is divided into 1~5 points based on the hydrogeological data.

Based on the scores of the five evaluation factors, ArcGIS (10.2, Esri, Beijing, China) was used to evaluate the applicability in the study area. Through spatial analyst of ArcGIS, the evaluation results were divided into five levels by their natural breaks (jenks): (1) unsuitable area, (2) small potential area, (3) general potential area, (4) medium potential area, and (5) high potential area. The zoning map for the adaptability zoning evaluation is obtained by ArcGIS in accordance with the spatial distribution of the recharge potential (Figure 9).

The results show that the western part of Liaocheng City is a suitable area for MAR. The overall score of the western region was higher because of the greater groundwater depth, thicker sand layer, and higher groundwater exploitation. Especially in Guanxian County, the groundwater depth was approximately 20 m, the thickness of the fine sand layer was the thickest, and the specific yield was the highest. This area is the high-potential area, while the medium-potential areas were near Linqing and Shenxian counties. The eastern part of Liaocheng City had a groundwater depth of 2 to 6 m. The thickness of the fine sand layer was relatively large, but the urgency of recharge was small. Therefore, the eastern part mainly had areas with less potential or that were unsuitable.

The main existing MAR project is the well–canal combination mode in Liaocheng City. The existing MAR project needs to be strengthened in the western region via water spreading methods, in which areas (4) and (5) are located at the end of the irrigation district of Liaocheng City. A new mode of the open channel–underground perforated pipe–shaft–water saving irrigation system should be extended in order to increase the groundwater recharge amount and expand the scope of groundwater recharge during the limited Yellow River water diversion period.

**Figure 8.** Maps of the five factors. (**a**) groundwater depth, (**b**) thickness of fine sand, (**c**) specific yield, (**d**) irrigation return flow, and (**e**) groundwater extraction intensity.

**Figure 9.** Adaptability zoning map of recharge and existing canal system in the Yellow River Irrigation District of Liaocheng City.

#### **4. Discussion and Conclusions**

In the 1970s and 1980s, various small-scale MAR projects were implemented to increase the amount of shallow groundwater in the North China Plain. Considering these characteristics and the relationship between surface water and shallow aquifers, MAR is divided into the three types: water spreading, well recharging, and a combination of both. MAR can be further divided into 10 forms according to the specific farmland water conservancy project (field infiltration, infiltration pond, infiltration ditch, ditch–underground permeable cement pipe–pond systems, tunnel–wells, seepage wells, shaft wells, canal–pipe–wells, brackish aquifer treatment, and ditch–well–check gate systems). These projects guarantee high grain yields and maintain the balance between recharging and extraction.

The Yellow River Irrigation District of Shandong Province is located in the lower reaches of the Yellow River. The effective irrigation area is mainly based on well irrigation. With the increase of crops, the water demand for agricultural irrigation has increased rapidly. Local groundwater can only provide half of the irrigation water, and Yellow River water diversion is performed to supplement irrigation. Large-scale overexploitation of shallow groundwater has occurred in some areas. The well–canal combination mode has been widely used in the Yellow River Irrigation District of Shandong Province for nearly 40 years. Because the well–canal mode has many advantages, it is applicable to the downstream, middle-stream, or upstream areas of the Yellow River Irrigation District. During water diversion, the water of the Yellow River is diverted by deep canals and relies on pumping irrigation and gravity drainage. As irrigation proceeds by pumping from canals, part of the water in the canals is recharged to the groundwater, which improves irrigation conditions. Without water diversion, the water demand can be satisfied by well irrigation. This mode has the advantage of recharging aquifers with river water and guaranteeing bumper harvests with wells. It also not only maintains high crop yields and basically guarantees the balance of exploitation and the recharging of shallow groundwater, but it solves the problem of aquifers being clogged by sediment from the Yellow River diversion because silt sediment at the canal head and sediment in the canal are dredged.

Based on field infiltration, infiltration ditches, and infiltration ponds, a new open channel–underground perforated pipe–shaft–water saving irrigation system was developed. The new system further expands the recharge scope and replenishment, and has three anti-blocking measures. The sustainable development of agriculture in the North China Plain is ensured by implementing the well–canal combination mode and adopting shallow groundwater recharge as the main line. The method of integrating MAR into agricultural facilities to form a farmland water conservancy system of water diversion, storage, infiltration, water savings, irrigation, and drainage is proposed to achieve the goal of comprehensively controlling droughts, floods, and salinization.

Liaocheng City was selected as the study area because of the distribution of aquifers, permeability of the unsaturated zone, and groundwater extraction intensity in the Yellow River Irrigation District of Shandong Province. An adaptability zoning evaluation system for water spreading was established based on the common modes of water spreading. Five factors were selected to reflect unsaturated zones and groundwater extraction: groundwater depth, thickness of fine sand, specific yield, irrigation return flow, and groundwater extraction intensity. The results show that MAR projects are adaptable to the western region and can resolve agricultural irrigation problems. The eastern and central regions have high groundwater tables, better diversion conditions for the Yellow River, and superior aquifer water storage capacities. However, these areas are not suitable for MAR projects due to their low groundwater extraction intensity. Thus, water diversion from the Yellow River and groundwater exploitation should be maintained in balance.

**Author Contributions:** This paper was composed by collaboration among all authors. Supervision, S.Q., Y.Z. and W.L.; Writing—original draft, S.L.; Writing—review & editing, W.W. All authors have read and agreed to the published version of the manuscript.

**Funding:** This study was supported by the Shandong Provincial Key Research and Development Project (2017GSF17121) and the Danish Development Agency (DANIDA) coordinated by the DANIDA Fellowship Center (DFC) through grant No. 17-M08-GEU.

**Acknowledgments:** The authors would like to acknowledge the editor and three anonymous reviewers for their valuable comments, which have greatly improved this paper.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

## *Article* **Evaluation of MAR Methods for Semi-Arid, Cold Regions**

#### **Nasanbayar Narantsogt \* and Ulf Mohrlok**

Department of Civil Engineering, Geo and Environmental Sciences, KIT—Karlsruhe Institute of Technology, 76131 Karlsruhe, Germany; ulf.mohrlok@kit.edu

**\*** Correspondence: ugeph@student.kit.edu or nnasan.4@gmail.com

Received: 22 August 2019; Accepted: 22 November 2019; Published: 2 December 2019

**Abstract:** Mongolia is a semi-arid, highly continental region with highly variable precipitation and river discharge. The groundwater aquifer located near Ulaanbaatar, the capital city of Mongolia, is the only one source for city water supply consumption, and it is important to ensure that groundwater is available now and in the future. The main watercourse near the capital city is the Tuul River, fed by precipitation in the Khentii Mountains. The semi-arid and cold environment shows high variability in precipitation and river discharge. However, due to absence of precipitation in winter and spring, the riverbed usually runs dry during these times of the year, and weather observations show that the dry period has been extending in recent years. However, in parallel with urban development, the extended groundwater aquifer has shown a clear decline, and the groundwater levels have dropped significantly. Therefore, a groundwater management system based on managed aquifer recharge is proposed, and a strategy to implement these measures in the Tuul River valley is presented in this paper. This strategy consists of the enhancement of natural recharge rates during the wet summer from the northern drainage canal, an additional increase in groundwater recharge through melting the ice storage in the dry period, as well as the construction of underground dams to accumulate groundwater and a surface water reservoir that releases a constant discharge in the outlet. To increase natural recharge rates of groundwater during the early dry period through the melting ice storage period, the MATLAB icing code, which was written for ice storage for limited and unlimited areas, was considered through finite element subsurface FLOW (FEFLOW) simulation scenarios as a water source in ice form on the surface. A study of the artificial permafrost of underground as an ice dam was processed in FEFLOW simulation scenarios for accumulating groundwater resources. The results of these artificial recharging methods were individually calculated, combined, and compared with the surface reservoir, which releases a constant discharge through the dam. In this paper, new ideas are presented involving managed aquifer recharge—MAR methods, and include application to aufeis, a mass of layered ice for groundwater recharge by melting. Additionally, the accumulation of groundwater using artificial permafrost is used as an underground dam. In addition, was considered recharging scenario only with constant release water amount from water reservoir also with all MAR methods together with reservoir combination.

**Keywords:** Ulaanbaatar; MAR; MATLAB; FEFLOW; artificial recharging scenarios

#### **1. Introduction**

The Tuul River, which flows through Mongolia's capital city Ulaanbaatar (UB), originates from the Khentii Mountains on the front side of the capital city. The water level of the Tuul River fluctuates according to an annual high-flow to low-flow cycle, with its average water flow of 26.6 m3/s [1].

This river also recharges the Ulaanbaatar and upper aquifers that provide the city's water supply by 218 wells divided into nine groups of intake wells and 21 booster pump stations operated by a

water supply agency. Daily domestic consumption fluctuates around 150–200 m3/day and depending on the four seasons; see Figure 1. The water supply agency estimates that another 130 <sup>×</sup> <sup>10</sup><sup>3</sup> <sup>m</sup>3/day of water is pumped from the aquifer in private wells by industries and individuals [2].

**Figure 1.** Groundwater aquifer south side of Ulaanbaatar city. (**a**). Hydrological drainage basins of Mongolia, (**b**). origin of Tuul River, (**c**). Ulaanbaatar aquifer and wells.

The joint meeting of the Mongolian and Russian mineral resources commissions estimated the useful capacity of groundwater resources of Ulaanbaatar aquifer at 264 <sup>×</sup> 10<sup>3</sup> m3/day [3], without an upper source in the eastern side of the Ulaanbaatar aquifer (Figure 1), which pump groundwater around from 24 <sup>×</sup> 103 <sup>m</sup>3/day to 48 <sup>×</sup> 103 <sup>m</sup>3/day from aquifer [4].

As with natural recharge formation of groundwater resources, the main components are water entering in the aquifer as result of surface runoff loss and the capacitive reserves of the upper water-bearing layer. During the period from May to December, the aquifer takes recharge due to surface runoff loss. In this period, replenishment infiltrates from the surface water to capacitive reserves of the stored groundwater in the Ulaanbaatar aquifer [5].

In March and April, the groundwater table of water source aquifers decreases to minimum of 8 to 10 m from the surface, but it reaches maximum level of 2 m under the earth surface in the central area of the source in June, July, and August [6], see Figure 2. This table begins from a hydrological year where the wet season begins with melt water from snow and ice after a long, cold winter.

Water demand increases day by day with the development of industries and increasing population growth, but groundwater resources have decreased due to the climate change and excess water usage. In the last decade, during spring (April and May), the dry period the Tuul River flow continues longer than one month after ice breakup [7].

**Figure 2.** Table of groundwater fluctuation depending on Tuul River flow.

The following Table 1 shows a calculated balance of the outflow and inflow at the Tuul River valley and Ulaanbaatar aquifer, which is used for the water supply of the city. The Tuul River and tributaries discharge was taken as the inflow rate, precipitation, and evaporation volume averaged over many years. The average discharge amounts over the last ten years of Tuul River was taken as inflow including other inflow sources as surface water from the Selbe, Uliastai, and Khul Rivers they flow to the Tuul River through aquifer for water supply of Ulaanbaatar city.


**Table 1.** Ulaanbaatar city water exploration volume balance for water supply.

The balance rate between surface water as inflow including precipitation and city consumption was estimated as following: from total inflow subtracted total pumped groundwater and added reused water from thermopower plants. see Table 1.

$$\mathbf{Q}\_{\text{balance}} = \mathbf{Q}\_{\text{infflow}} - \mathbf{Q}\_{\text{pumped}} + \mathbf{Q}\_{\text{reused}}$$

The above estimated calculation is based on the total water supply of the upper and central sources of Ulaanbaatar aquifers with the annual average flow rate of the Tuul River.

The water consumption of the whole aquifer of Ulaanbaatar, including domestic and thermal power plants, is shown above and is sorted by hydrologic year from April, when balance starts with a positive value [8].

The total amount of water consumption in Ulaanbaatar used for water supply in the dry, non-recharging season from surface water of Tuul River reaches about <sup>−</sup>20 <sup>×</sup> 106 m3/year depending on the weather conditions and general precipitation in a given year. This balance between water consumption and recharging inflow from Tuul River was estimated with all pumped groundwater from all nine water sources, including an upper source.

The balance between incoming flow coming from the Tuul River to the aquifer and outcoming flow as an exploration rate by intake wells [9]. This means city consumes <sup>−</sup>20 <sup>×</sup> 106 m3/year groundwater

from aquifer without recharging from river. The balance estimation between groundwater income and pumped water for domestic water supply from the Ulaanbaatar aquifer source area without thermopower plants water supply consumption is <sup>−</sup>7.1 <sup>×</sup> 106 <sup>m</sup>3/year water.

Therefore, there is an urgent need to solve this problem through the design of hydraulic structures for an underground dam, managed aquifer recharge, and the promotion of ice storage or the building of a surface reservoir dam that releases constant controlled outlet water.

In the next chapter, this problem is considered only for domestic consumption in the central source area of the Ulaanbaatar aquifer. A finite element subsurface FLOW (FEFLOW) simulation was taken in the upper and central source areas of the A zone, where there are only 23 wells for water supply; see Figure 3.

**Figure 3.** Ulaanbaatar aquifer, central source area, and the simulated A zone area.

The average date of the first ice formation on rivers is the third week of October. The freezing of the rivers starts from the end of October and lasts until the end of December. The ice cover lasts, on average, for 145 days. During the last 60 years, the annual mean of air temperature in Mongolia has increased by 1.66 ◦C, the winter temperature has increased by 3.61 ◦C, the spring–autumn temperature has increased by 1.4–1.5 ◦C, and the summer temperature has had no clear trend [10]. Temperature has rapidly increased in March, May, September, and November, and the ice regimes of the Mongolian rivers have therefore changed [2–4].

Ice phenology has shifted about 3–30 days in terms of freeze-up and break-up dates, and ice cover duration has shortened. Maximum ice thickness also decreased from the 1960s to 2000 [11].

A small river like Tuul is frozen to the bed for 2.5 months [3], and in mid-April, it has no ice cover with some places being dry bed without flow.

#### **2. Materials and Methods**

#### *2.1. FEFLOW (Finite Element Subsurface FLOW and Transport System) Transit Model with Ice Scenarios*

Nowadays, groundwater simulation and modeling are some of the main tools for groundwater aquifers [12], which visualize the situation and conditions of water in underground porous media for the protection of groundwater, as well as the restoration and development of aquifers. Groundwater level loggers collect more data, but because they cannot be manually controlled, only software modeling and processing provide an option that is both fast and accurate [13].

Every spring over the last decade, during the months of March and April, the Tuul River has dried out or has not flowed [14]. Therefore, we need to address on this problem by building complexes of hydraulic structures, establishing measures for flow control, and building artificial groundwater recharge systems like drainage or flooding areas near drinking water extraction wells. This study attempts to find suitable artificial groundwater recharging methods for the upper part of the central source groundwater aquifer, encompassing the water supply source area of the Tuul River valley inside Ulaanbaatar. The central source of drinking water supply system was established and put into operation in 1959. Drinking water is extracted from 93 deep well pumps with seven booster pumping stations; see Figure 4.

**Figure 4.** Table of groundwater fluctuation in Wells 51 and 68 in the simulation area.

The capacity of the source reserve is 114 <sup>×</sup> 103 m3/day. Nowadays, water extraction is from 70–80 wells, and its volume reaches 87–90 <sup>×</sup> <sup>10</sup><sup>3</sup> m3/day for supply to the capital city [4]; see Figure 3.

The central source of the A–A zone was simulated in FEFLOW simulation. The upper part of the central source intake area near Well 68 takes recharge from river surface water until January, when the Tuul River flows under ice cover. After that period, the Tuul River freezes while the riverbed bottom and groundwater continuously decrease until May. However, in the western side of Zone A, the intake wells area near Well 51 recharge comes not only from Tuul River but also from the Uliastai River; therefore, recharge takes place early in May. As shown in graph, the Uliastai River flow freezes early in November as a small river due to a groundwater decrease that begins in October, and the following recharge in May depends on melting water from both rivers; see Figure 5.

**Figure 5.** Icing dynamic process in the Uliastai River valley.

The groundwater recharging process runs as follows:

In the initial phase in which surface runoff is non-existent, maximum groundwater declines are observed from the end of April to early May. If the value of the groundwater table is same and above the elevation of the riverbed, it becomes possible for the river to flow downstream without recharge.

The icing phenomena is one of the most important parameters for semi-arid, highly continental climate conditions and plays significant role in the hydrological cycle and regime of Mongolia. The icing or ice cover of rivers and lakes takes place in the cold season, which occurs over five to six months, when the ice cover thickness reaches to 0.8–3.2 m. However, big mountain rivers with a greater slope and bigger perturbation boulders in riffle sections stay open for the whole year and do not completely freeze along the length.

Generally, for surface ice cover to become firmly established, the mean (depth-averaged) temperature of water must be less than 2 ◦C, the daily average temperature must be less than −5 ◦C [15], and the wind speed must be less than 5 m/s [11].

The Uliastai River, one of the tributaries of the Tuul River, originates from the Khentii Mountains, flows from the north east side to the south through Ulaanbaatar, and contributes to the Tuul River through the water supply wells of the central source for Ulaanbaatar.

Icing in the aufeis accumulates during winter along the streams and the river valley in northern Mongolia, which is dominated by semi-arid, highly continental regions. In the Uliastai River, building ice lasts from the middle of October until the end of December, and the melting process starts from the end of March and ends in April.

The icing dynamics depend on the groundwater fluxes and discharges alongside of the main channel. The spring leakes through drainage channel, builds ice sheets over the frozen riverbed, where the main stream flows under ice cover. The ice generating process and icing dynamics are studied from the middle of October to the end of December.

The main stream flows by the main channel under ice cover. The side spring streams over the top of the ice sheets. This phenomenon is called icing or aufeis. Icing or aufeis consist of sheets of stratified ice formed by freezing consecutive water leaks [16].

The water flows over existing ice layers. It forms through the upwelling of groundwater discharge or manmade drainage channels, where groundwater discharge is blocked by ice, perturbing the steady-state condition and causing a small incremental rise in the local water table until discharge occurs along the bank and over the top of the previously formed ice [16].

In the beginning of November, river flow freezes from side benches, with the spring discharge no longer extending energy because it had frozen first. After the riverbed completely freezes and takes ice cover, the spring discharges leak from under the ice or ice hummock while the groundwater head and pressure flows increase over frozen ice sheets to create the next ice sheet.

A groundwater flux from drainage canal flows on surface as spring and creates the next ice sheet, which fills the lower ravines and smoothest horizontal by ice sheet.

The groundwater flux alongside the river drains through drainage canal-built icing phenomena, while springs' flow beds are blocked by ice and the main stream in the main riverbed flows under the ice cover. In this section, the Uliastai River gains streams of almost one-third of the flow of the discharge abstracts to groundwater, and groundwater comes from springs.

Thus, phenomena create side spring leakage from under frozen soil, leakage which flows over frozen soil and ice-covered streams, fills ravines and lower lands, creates ice sheets on the ice until the river valley gains the same level of ice. After that, the average daily temperature decreases to under −20 ◦C, and then the groundwater leakage discharge decreases the head and pressure of groundwater flux in underground flows. Additionally, drained groundwater, spring discharge from the surface, and groundwater decreases influence the quantity of groundwater volume. The ice thickness measurements from the 15th to the 30th of December showed that the thickness of the aufeis sheets have not increased; see Figure 4.

In this way, groundwater leakage, like drained water or springs over frozen soil and ice cover, creates stratified ice sheets over other ice sheets. In the Uliastai River, there are spring discharges of 32

L/s—a small amount of water that nonetheless build ice thicknesses in some places up to 1 m thick. Some of the rivers make ice sheets of thicknesses of several meters. The decline of the ice storage depends on the quantity of spring discharge. The end of building aufeis is usually in the last days of December or the first days of January—after the longest December night ends and the colder days of the year start. In a larger river, such as the Tuul, is possible to create an aufeis until February. An aufeis typically starts to melt during summer and finishes by the end of April (or sometimes, the beginning of May), and it will often form in the same place each year [7].

From these icing ideas, a MATLAB code was written and used for the management of aquifer recharge for the central source of the A zone, which accumulates ice storage due to the use of melt water for recharge in the dry season [17].

The aufeis code was written as one mole of water exchanging energy with cold air and ice, extending velocity of in both the x and y directions could widen in an ellipsoidal way due to the Darcy–Weisbach law. The aufeis spreading dynamics of one mole thick water sheet over ice sheets decrease the extension size while decreasing air temperature day by day from −5 to −30 ◦C [17].

The result shows that length of ice spread is 2329 m, with a width of 458 m and thickness of 1.54 m. The code for ice storage was simplified with the same slope along flat area. The MATLAB code presents the ice extending dynamics of spreading water with pipeline levees, and from leakage point to pipeline levee of 1500 m. The length of ice storage is only 1500 m; see Figure 6. The results show that the length of ice spread is 1500 m, with a width of 400 m and thickness of 2.72 m, thicker than the unlimited area. In the following figure, you can see two ice storages: unlimited and unlimited area with underground pipeline levee, see Figure 7. These ice storages will change recharging boundary condition in FEFLOW model as a one-month early recharge in April by melt water from north side, where the Tuul River flows in May [8].

The relation between the estimated quantity of the MATLAB ice code and the FEFLOW simulation characterizes the boundary condition change in the northwest side of simulated area. The groundwater recharge from the river starts in May, when the Tuul River flows again. However, in the simulation, the aufeis-stored ice was shown to melt earlier in April and to recharge groundwater. The aufeis changes the northwest boundary condition so that the recharge starts from the beginning of April. The drainage canal begins to flow with water from May, when the Tuul River flows again; see Figure 6. The eastern boundary groundwater fluctuation taken from Well 68 was not found to change. However, the downwards flow direction of the northwest boundary condition was found to change with recharge from the ice melt water; see Figure 6, blue line.

**Figure 6.** Recharging time difference on northwest boundary (NWBC) with ice storage (blue), only from drainage canal (red).

**Figure 7.** The water supply pipelines connecting wells.

One of the artificial recharging groundwater resources is the temporary and spatial redistribution of surface runoff, which in the beginning of winter results in ice formation. The surface runoff flows over the northern side of the central section A zone through filtration channel will release water at the end of canal and, thus, it again creates more ice sheets over frozen ones.

There are three ways of promoting ice creation in cold regions in winter: on the ground surface, in the underground open pit or channel, and on the river bed. Of these, surface and underground ice creation were considered in the FEFLOW simulation scenarios. See Figure 7.

The rate of water release is 1 m3/s at the beginning of November until the middle of December, and creates ice storage when the average temperature decreased under −5 ◦C.

From the above figures, it is shown that the northern drainage canal allows us to accumulate 1 m3/s flow water for a month, from November to the middle of December, and 3.9 <sup>×</sup> 106 <sup>m</sup><sup>3</sup> water can be stored on the surface. However, with losses from evaporation and winter fog over frozen ice sheets during the melting season, evaporation loss allowed only half of this quantity—about 2 <sup>×</sup> <sup>10</sup><sup>6</sup> m3—to accumulate for recharging groundwater. Here melted water from ice storage recharges groundwater in the central source A–A zone from the northern side, from the beginning of April until May.

Ice accumulation from November to the end of December also recharges groundwater while water flows through the drainage canal. It is then transferred to the artificial regime with the subsequent supply of water to canals, functioning together until the period of ice formation, provided there is ice accumulation from surface water on the end of the canal.

The underground dam will be built in the first layer until reaching the natural permafrost layer 5–10 m from the surface [18,19], considered as an ice wall on the western boundary line.

#### *2.2. FEFLOW Steady Model with Water Reservoir Scenario*

A preliminary analysis, as presented in this study, was to identify low-cost MAR implementation measures adapted to the specific natural conditions of the Northern Mongolia. Thus, the coldness of cold weather can be used to keep water in ice form and as a water resource in the winter season in addition to being used during the dry season, characterized by low flow, by melting ice when the rivers have dried out.

The accumulated ice would recharge the groundwater in the dry season from March to May by melting, and the riverbed would be dry without water or cover ice.

To find the total additional water resource recharged by the northern drainage canal and ice storage, the abstraction rate should be increased until the available maximum rate, when groundwater drawdown decreases under the filter screen at the bottom of wells and soaks up air. For calculation of the potential maximum abstraction quantity in the central source A zone without MAR methods, it should be estimated using FEFLOW simulation until some well groundwater drawdown reaches to the bottom of the well screen, pumping air instead of water. In this way, groundwater fluctuation ranges can be established in FEFLOW.

The groundwater fluctuation graph of monitoring well N8 show that by increasing daily extraction to 60,986 m3/day, it was not possible to pump water while the groundwater drawdown was under the bottom screen. However, in the middle of group of wells, groundwater level is in extreme drawdown, and only 45,121 m3/day is possible. Therefore, the maximum abstraction rate of this area is 45,121 m3/day, which means 16,469,165 m3/year <sup>=</sup> 16.5 <sup>×</sup> 106 <sup>m</sup><sup>3</sup> of water per year.

Nowadays, the exploitation rate is 20,329 m3/day (average of 2009–2011), 7.5 <sup>×</sup> 10<sup>6</sup> m3/year corresponding to half of the possible maximum abstraction.

The recharge quantity with MAR method and then the maximum abstraction rate increase until 70,133 m3/day, which corresponds to 25.6 <sup>×</sup> 106 m3/year.

The difference between the above maximum abstraction simulations shows that it is possible to increase the groundwater resource extraction rate to around 25 <sup>×</sup> 103 m3/day using MAR methods. That means approximately 912,500 m3/year = 9.125 <sup>×</sup> 106 m3 additional groundwater during wet season could be kept as reserve in the upper part of this central source area A–A zone.

The percentage for recharge groundwater sources in this area was estimated by the sum of all methods of each MAR method, and is taken as 100% for all together.

The percentage of the increased amount of water reserve by each method, after combination of the MAR methods, from increased groundwater sources is as follows.


All of these FEFLOW simulations were simulated in a transient model, where the river surface water level increases in the wet season over the riverbed, and decreases in the dry season under the dry riverbed. The table of the eastern and western boundary groundwater fluctuates under the surface, depending on the recharge from the river for whole year.

Another alternative simulation scenario is for the water release from the surface water reservoir with a dam, with a constant q rate of 26.6 m3/s and creation of table for constant groundwater water. In this case, the model is taken with the eastern and western boundary using a constant groundwater level and hydraulic head and, also, in the southern boundary as the Tuul River with a constant level value for groundwater recharge. In the FEFLOW simulation, the maximum abstraction rate reached 90,288 m3/day. Following this, the simulation produced an error report indicating that the aquifer has no groundwater; see Figure 8.

**Figure 8.** The groundwater contours and simulation error by pumped 166,700 m3/day water.

This means the flow control by surface reservoir accumulates flood water and, additionally, the recharge groundwater quantity for the central source of A zone is 16.5 <sup>×</sup> 106 <sup>m</sup>3/year.

The flow control upper reservoir releases a constant outflow throughout the year, and it means it also recharges the central source of B zone, occupying a larger area twice in size, with 50 intake wells. The constant release rate for the whole central source area that is recharged from the reservoir would be 49.5 <sup>×</sup> 106 m3/year. An additional 350 <sup>×</sup> 106 m3 of accumulated water in the reservoir would be reserved for use as a freshwater resource [20].

All these simulated calculations involved recharging water from only reservoir. When we are simulating a combination of artificial recharging groundwater sources and reservoir release for surface water recharge, then the additional recharged groundwater would be 167,700 m3/day; see Figure 9.

**Figure 9.** The groundwater simulated contours by a combination of managed aquifer methods and recharge from Tuul River outflowing from reservoir.

These MAR methods and the reservoir recharge system include the following simulation scenarios.


In this combination of scenarios, the maximum possible abstraction rate is increased up to 167,700 m3/day after simulation. The maximum possible abstraction quantity per year reaches up to 44.7 <sup>×</sup> <sup>10</sup><sup>6</sup> m3/year.

The FEFLOW simulation was only run for the central source of A zone. When we take the second B zone with 50 wells, this maximum abstraction rate would be increased, and at least doubled or 89.5 <sup>×</sup> <sup>10</sup><sup>6</sup> <sup>m</sup>3/year due to water recharging because the B zone is twice as large as the A zone. If the Tuul River flows yearly with constant discharge, then river surface water will be recharged for industrial sources, and that leads to increases in the additional groundwater resource. This will increase the flow from groundwater sources to 149 <sup>×</sup> <sup>10</sup><sup>6</sup> m3/year and create natural conditions for discontinuous Tuul River flow to assist in avoiding a situation of shortages in the fresh water supply for Ulaanbaatar city.

#### **3. Analyses of FEFLOW Model**

Aufeis is an icing method that brings more recharge water in the middle of the study area for its other constituent subsections as well as an ice wall, increasing the backside of the western boundary. The drainage canal filtrates through the northern side and the recharge occupies the whole area from the east to the west boundary.

The combination of all methods includes an ice wall, icing, and drainage canal. The drainage canal filtrates water from east to west along the north side of the study area, and the rest are released from it as water, since early winter creates icing, see Figure 10. Thus, this icing method is performed after drainage canal filtration has ceased altogether.

As seen in the following figure, recharge from the drainage canal begins in May and ends in November, with icing recharge beginning in the middle of March and lasting until May, and the ice wall holding groundwater undergoing the reverse, from March until November.

In nature, we can reserve more water in ice form as aufeis to help in building some hydraulic structures, such as drainage canals or underground dams. Both of these structures can help us to reserve water stores underground, as well as keep them on the surface in ice form.

From October to December, surface water naturally accumulates and is kept in ice form from river flow in the semi-arid, highly continental region of the study area. During the subsequent dry season, these sources increase and recharge potential water availability by melting water sources.

One possibility is to reserve water resources to regulate groundwater flux control in highly continental cold regions to eliminate dry riverbeds and keep primary source rivers, such as the Tuul River, continuously flowing during low flow periods is by melting ice blocks.

From the following graph, it can be seen that the northern drainage canal will allow accumulation 1 m3/s flow water for a month from the beginning of November to mid-December, creating 3.9 <sup>×</sup> 106 m3 water for surface storage.

However, due to losses from evaporation and winter fog over frozen ice sheets and, also, evaporation loss during the melting season, only half of this quantity, or about 2 <sup>×</sup> <sup>10</sup><sup>6</sup> m3, will actually accumulate and be available for groundwater recharge.

**Figure 10.** The combination of all variations in simulated area compared to the measured groundwater level.

For this FEFLOW simulation, the maximum difference between the simulation in natural conditions and the simulation in each managed aquifer recharge scenario is taken into account. The result of the FEFLOW simulation in each MAR method scenario demonstrates the following results:


#### **4. Results and Conclusions**

In recent decades, as the city develops and expands, the consumption of the domestic and industrial water supply of the city is increasing intensively, but the availability of the water supply, both now and in the future, has become a pressing issue. Therefore, in this paper, possible variants for recharging the groundwater resource are considered using FEFLOW simulation, and compared with the results for modeling in the upper part of Ulaanbaatar aquifer.

The simulation variants of the recharging methods of managed aquifer are northern drainage canal, which recharges groundwater from the opposite side such as for the Tuul River, icing or aufeis, which keeps water in ice form, brings it from the end of the wet season through winter to the dry season, and increases groundwater sources, by melting. In the end of the upper part of the aquifer in the western side, building an underground dam which accumulates groundwater on the backside. All these variants were used in FEFLOW simulation, and the recharged quantity was calculated separately, and combined ones demonstrated the following results.

• A single ice wall on the western boundary in natural conditions without drainage of canal increases groundwater resources by 516 <sup>×</sup> 103 m3/year.


The additional recharged groundwater difference by increased daily abstraction rate from this aquifer both with and without managed aquifer recharge would be 9.16 <sup>×</sup> 106 <sup>m</sup>3/year.

When the FEFLOW simulation scenario has taken into account the surface water reservoir, which releases constant discharge flow, then the calculations involving recharge from only reservoir separately and also the combined simulation with MAR methods demonstrate the following results.


These simulation results are from the upper part with only 23 wells of central source area, which has a total of 72 wells. When the simulation is based on all the Ulaanbaatar aquifers and additional recharged groundwater resource from Tuul River flow with constant runoff, the results is at least 3 times more than this small area or 150 <sup>×</sup> 106 m3/year.

The simulation of both transient (MAR methods) and steady reservoir outflow recharge methods show that artificial recharge methods of groundwater are not enough for additionally recharged quantity compared with surface reservoir outflow recharging. However, a combination of both methods demonstrates that use of artificial recharging groundwater sources in the aquifer is more effective.

Using coldness and icing in a semi-arid, highly continental region such as Mongolia can increase groundwater resources. The nonconventional artificial recharging methods specified in this paper, including ice storage, can assist in groundwater recharge in dry seasons, when there is no flow and dry beds in river systems.

The Tuul River reservoir will be located upstream from the aquifer and can accumulate 350 <sup>×</sup> <sup>10</sup><sup>6</sup> m3 in water [20], creating a more humid environment in the region of Ulaanbaatar. A positive side effect of increased evaporation from the open water is that air humidity will be increased, which will lead to artificial precipitation and deposition of atmospheric particles from air pollution and, thus, helps to improve air quality in the city and its surroundings.

Groundwater flux disturbed by the presence of an ice wall comes out on the surface during spring, spreading, evaporating, and losing some of its mass. This creates fog in winter and turbidity in air when a cold high-pressure cyclone dominates Ulaanbaatar city. The turbidity of air corresponds to windy surroundings and clears air pollution.

Following the MAR method, water storage in the form of ice in semi-arid, cold, highly continental regions would also produce coldness during the dry season and increase the inner continental hydrologic cycle through melting and evaporation. In the dry season, melting ice evaporates and increases precipitation. Many forest fires occur every year in Mongolia during the dry season due, at least partially, to the lack of precipitation and the lower air humidity. Therefore, ice-keeping methods would help in keeping the environment green, environmentally close to the natural process and improve human and natural habitats.

The aufeis blocks, involving icing in the river valley, are increasing continental cycles by evaporating melted water and producing coldness [21]. Thus, it helps in growing of vegetation and increasing precipitation in the dry period. During the fall months, from September to October, the soil moisture increases with more rain, and during spring (March and April), there is more humidity in the air and more rain. Meltwater is a common source for rivers and enables their continuous flowing [22] as well as the recharging of groundwater aquifers.

**Author Contributions:** U.M., First editor and reviewer of my (N.N.) dissertation structure. This article structured by his (U.M.) recommendation which is part of my (N.N.) dissertation in groundwater modelling and management.

**Funding:** This research received no external funding.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

## *Article* **New Methods for Microbiological Monitoring at Riverbank Filtration Sites**

#### **Yasmin Adomat 1,\*, Gerit-Hartmut Orzechowski 1, Marc Pelger 1, Robert Haas 2, Rico Bartak 2, Zsuzsanna Ágnes Nagy-Kovács 3, Joep Appels <sup>4</sup> and Thomas Grischek <sup>1</sup>**


Received: 19 November 2019; Accepted: 18 February 2020; Published: 20 February 2020

**Abstract:** Water suppliers aim to achieve microbiological stability throughout their supply system by regular monitoring of water quality. Monitoring temporal biomass dynamics at high frequency is time consuming due to the labor-intensive nature and limitations of conventional, cultivation-based detection methods. The goal of this study was to assess the value of new rapid monitoring methods for quantifying and characterizing dynamic fluctuations in bacterial biomass. Using flow cytometry and two precise enzymatic detection methods, bacterial biomass-related parameters were monitored at three riverbank filtration sites. Additionally, the treatment capacity of an ultrafiltration pilot plant was researched using online flow-cytometry. The results provide insights into microbiological quality of treated water and emphasize the value of rapid, easy and sensitive alternatives to traditional bacterial monitoring techniques.

**Keywords:** online flow-cytometry; enzymatic activity; riverbank filtration; ultrafiltration; ATP

#### **1. Introduction**

Riverbank filtration (RBF) systems are operated in many countries for the public and industrial water supply due to their efficient removal of pollutants such as microorganisms [1,2]. During RBF, the removal of bacteria, viruses, and protozoa in surface water is attained trough filtration, sorption, and grazing processes besides die-off. Such processes can be influenced by the aquifer material composition, hydraulic gradient, temperature, redox conditions, organic/inorganic nutrients, and travel time in the aquifer [3–5].

By measuring the microbiological characteristics such as bacterial biomass concentration or enzymatic activities, these interactions between surface water and groundwater can be determined. The access of limited nutrient supply (e.g., organic carbon, nitrogen or phosphorus) or environmental conditions (temperature, inhibitory substances) can lead to complex interactions between various microbes [6]. As a result, unwanted changes in microbiological water quality such as an excessive growth of bacteria can lead to a degradation of drinking-water quality and operational problems [7,8].

The World Health Organization (WHO) specified that water which enters the distribution system must be microbiologically safe and ideally should also be biologically stable, meaning microbiological water quality must be maintained from the point of drinking water production up to the point of consumption [9,10]. To ensure safe and effective water treatment, distribution, and consumption, reliable procedures for characterizing and monitoring waterborne microbes need to be carried out by water suppliers on a regular basis.

Water produced at RBF sites is commonly monitored for the absence of pathogen indicator organisms like *Escherichia coli (E.coli)*, total coliforms (TC), enterococci and *Clostridium* using cultivation-based methods with targeted growth media [4]. Additionally, heterotrophic plate counts (HPC) with non-specific media are frequently assessed. Although there is no evidence of a link between HPC results and health risk, it is of major importance to assess data of microbiological growth during drinking-water treatment, and to detect changes in bacterial concentration and composition of monitored water [5,6,11]. Since the HPC method was first introduced in the 1800s as a public health indicator, science has advanced. Hence, HPC monitoring became more useful as an operational rather than a health-based indicator [9]. At present, within water-treatment facilities in Germany, The Netherlands, and Hungary the HPC method is used for validation and verification of drinking-water treatment processes. Abnormal changes in HPC indicate problems in the treatment process and appropriate actions are essential to ensure that the problem is identified and eleminated [9].

Although the HPC method was introduced more than 100 years ago and enhancements regarding general performance and interpretation of data were developed, additional work concerning sample incubation times, temperatures, and acceptable critical thresholds is required [12]. Despite time-intensive laboratory procedures and incubation, the HPC method only detects a fraction of bacterial cells in water samples. This is due to the fact that only 0.1%–1.0% of bacteria species present in aquatic samples is culturable under laboratory conditions which was confirmed in various studies [5,13,14]. Furthermore, an estimation of the percentage of subpopulation of heterotrophic bacteria as well as a differentiation of which of these subpopulations include potential pathogens is not possible using HPC techniques [5].

In the past decade, significant advances in rapid cultivation-independent techniques, mostly fluorescence-based methods, have been developed. These methods focus on direct measurements of indigenous bacterial growth or enzymatic activity. Examples include optical methods (e.g., flow-cytometry, FCM) which count suspended particles in water samples and are able to differentiate between bacteria and abiotic particles based on e.g., 3D scanning or chemical staining techniques (e.g., SYBR®Green or propidiumiodide) [14,15]. Also, on-site sensors measuring indirect indicators of microbiological fluctuations such as adenosine triphosphate (ATP) concentration or specific enzymatic activities have been developed in the past decade [16–18].

In this study, FCM and two enzymatic detection methods (ALP, *alkaline phosphatase*) were used to analyze the total microbiological water quality after different treatment processes in samples of three RBF treatment sites. The two main objectives were to assess the applicability of each method in routine monitoring programs and, to compare the methods with each other as well as with HPC data. Additionally, an ultrafiltration pilot plant was monitored using online FCM with the goal to assess the pilot plant´s performance and to test if a continuous measurement of bacterial removal is possible.

#### **2. Materials and Methods**

#### *2.1. Site Description*

Samples were collected from continuously and discontinuously operated sample taps from two RBF sites around Dresden: D1 and D2, Csepel Island, Hungary (CI), and various RBF wells at Szentendre Island (SI), Hungary. D1 and D2 are situated on the floodplain of the Elbe River around Dresden, the state capital of Saxony, Germany. D1 has a total capacity of 72,000 m3/day, 111 vertical siphon wells as well as 36 single-operated wells [19]. The waterworks operates two separate treatment trains: a RBF treatment train and a managed aquifer recharge (MAR) treatment train. After well extraction, the RBF and MAR water is aerated and filtered using granular activated carbon (GAC) and disinfected with chlorine before it is distributed as drinking water. Water in D2 is abstracted from three siphon well galleries with 72 vertical wells with a total capacity of 36,000 m3/day (approx. 65%–80% bank filtrate) and post-treated by cascade aeration, GAC filtration and disinfection with chlorine dioxide. In Hungary, 756 horizontal as well as vertical wells on CI and SI are operated by Budapest Waterworks Ltd with a maximum capacity of 1.0 million m3/day and an average supply of about 456,000 m3/day [20]. Post-treatment is performed in Cl using ventilation, coagulation (aluminium sulphate), ozonation, sand filtration, ultraviolet (UV) radiation (temporary), and disinfection with chlorine.

#### *2.2. Sample Collection and Microbiological Characterization*

Prior to sampling at discontinuously operated taps, a disinfection-step with ethanol (96% Merck KGaA, Darmstadt, Germany) and flame sterilization (propane/butane gas, 1350 ◦C) were applied followed by a 3 min flushing interval before samples were collected. Samples were collected into 5 mL (polypropylene rack tube, Corning, New-York, USA), 15 mL (polypropylene rack tube, VWR, Radnor, USA), and 1 L (Borosilicate glass with polypropylene screw cap, VWR; Radnor USA) sterile sample bottles. Afterwards, each sample was transported to the laboratory for analysis within 12 h.

Intracellular ATP (ATPi) was determined using a luminesce-based Clean-Trace ATP water test kit (3M, St. Paul USA). Based on an enzymatic reaction (firefly luciferase), total (ATPt) and extracellular ATP (ATPe) were measured in relative light units (RLU) as per the manufacturer's instructions. ATPi was calculated from ATPt and ATPe values. Using a fully automatic operating system called BACTcontrol (microLAN, Waalwijk, Netherlands), total enzymatic activity (TEA) was analyzed by measuring the specific activity of ALP as an indicator for the presence of bacteria. Prior to each measurement, the water sample was pumped at 0.2–1 mL/s through a 0.45 μm ceramic filter into a reactor chamber. While constantly stirring, the concentrated water sample was incubated for 20 min at 45 ± 0.1 ◦C. During incubation, the enzymatic activity of ALP was detected in methylumbelliferone (MUF, in pmol MUF/(min·100 mL) by a fluorimeter which was pre-calibrated using a standard concentration of 1000 nM MUF.

Flow-cytometry analysis was carried out with a BactoSense (Sigrist, Switzerland) flow-cytometer equipped with a 488 nm solid-state laser and an optional continuous/discontinuous sampling port. Sample volumes of 260 μL were drawn at a flow rate of 200–400 mL/min and mixed with fluorescent stain (SYBR®Green, propidiumiodide). After incubation (10 min, 37 ◦C), samples were analyzed (FL1 channel at 525 nm, FL3 channel at 721 nm) using fixed gates to separate cells and background signals and additionally to distinguish between so-called high (HNAP) and low (LNAP) nucleic acid content cells.

#### *2.3. Ultrafiltration Pilot Plant and Online Flow-Cytometry (FCM) Measurements*

The ultrafiltration pilot plant was operated at D1 with a treatment capacity of 20 m3/h. Using either Elbe River water or flocculated Elbe River water as feed supply, water was pumped directly into a storage tank (1.9 m3) through a supply pipe. Ultrafiltration was processed by using two membrane modules consisting of polyvinylidene difluoride (PVDF) with pore sizes of 20 nm (UF 1, Pall Corporation, Port Washington, USA) and 18 nm (UF 2, inge GmbH, Greifenberg, Germany) and operated at a flux of 40–80 L/(m2 h) at 1.5 bar.

Online FCM sampling was realized using an automatic programmed magnetic-valve system (Figure 1) consisting of three sampling ports (Feed, Permeate 1, and Permeate 2). Water samples were drawn in a bypass which was controlled using a programmed Delphin-EMS-system (Delphin Technology AG, Gladbach, Germany) and the online sampling setting of the BactoSense flow-cytometer. Sample analysis was carried out applying the same method which is outlined in Section 2.2.

The cleaning process of the membrane modules was adjusted every 30 min by backwashing. UF 1 was cleaned using a combination of air and water at a flux of 5 m3/h for 60 s. While backwashing air was added in the membrane direction, water was added in the reverse direction. Afterwards, a 45 s forward flush processing step in the direction of filtration at 7 m3/h was performed. UF 2 was pre-cleaned by air flushing for 10 s followed by air–water backwashing in the flow direction for 50 s. The cleaning was completed with a 15 s forward flush step at 7 m3/h. Taking into account backwash

cycles every 30 min to avoid a sampling of backwash water, samples were drawn by flushing the FCM system for 2 min and analyzed within a cycle of 105 min according to Table 1.

**Figure 1.** Scheme of online flow-cytometry (FCM) sampling system.



#### *2.4. Data Analysis*

Statistical data processing was carried out using MS Excel and OriginLab. All microbiological data analysis were carried out using the provided device-specific software.

#### **3. Results**

#### *3.1. Correlation of New Methods and Conventional Cultivation-Based Methods during Riverbank Filtration (RBF) and Drinking-Water Treatment*

To elucidate the correlation of FCM, ALP-TEA and ATP values a serial dilution was applied with ultrapure water (autoclaved at 121 ◦C and 15 min, V75, SysTech, Pegnitz, Germany) and bottled mineral water (Evian, France). FCM and ALP-TEA provided positive bacterial counts for all water samples with an average maximum Pearson correlation coefficient of R = 0.80 for the intact cell count (ICC). There was a high correlation between ICC and ALP-TEA (Table 2). This was also observed for the total cell count (TCC)/ATPt ratio (R = 0.78) and the ALP-TEA/HNAP (R = 0.76).

**Table 2.** Correlation coefficients of FCM, ALP-TEA and ATP methods.


The sample that contained 5% bottled drinking water showed a slight increase by 16.3% ICC with FCM and ATPt which is not in accordance with the ALP-TEA decrease by 18.3 pmol/min (Figure 2). Also, ALP-TEA/ATP standard deviation in 100%-ultrapure water are fluctuating around ± 10 pmol/min which may be caused either by background noise or still intact cells with a high amount of intercellular ALP. Suprisingly, a higher correlation coefficient with ALP-TEA/ATPt (R = 0.57) than ALP-TEA/ATPi (R = 0.32) was determined given the fact that the ALP-TEA amount refers only to living organisms and, therefore, a better ALP-TEA/ATPi would have been expected. Further to note is the higher correlation of ATPi/HNAP (R = 0.87) than ATPi/ICC (R = 0.76) ratio. These values are in accordance to [21] since HNAP refers to the ratio of large cells to small cells and could therefore contain higher levels of intracellular ATP.

**Figure 2.** Correlation of FCM (n = 5), total adenosine triphosphate (ATPt, n = 3) and alkaline phosphatase-total enzymatic activity (ALP-TEA, n = 3) for measuring total biomass related units (BRU), samples from bottled mineral water (Evian) were diluted with ultrapure water from the same bottle, based on a method described in [13]. FCM samples were stained with SYBER®Green and propidiumiodide, error bars indicate standard deviation on samples.

To further demonstrate potential links between FCM and ALP-TEA related data, results of the CI, D2 and D1 sampling campaigns are given in Figure 3. Unfortunately, ALP-TEA sampling was only possible in CI due to logistical problems in D1 and D2. 1.5 <sup>×</sup> 102 ICC/μL and 20.2% HNAP were detected in RBF samples with corresponding ALP-TEA values of 70 pmol/min in CI. In D2 and D1 similar values of 3.7 <sup>×</sup> 10<sup>2</sup> ICC/μL (D1) and 1.4 <sup>×</sup> 102 ICC/μL (D2) were measured in RBF samples with an average HNAP portion of 20.9% and 20.4%. Regarding Elbe River water and RBF, BRU in D1 decreased by 97.0% (1.53 log units).

Biomass concentration increased further in open aeration towers by 4%, to 1.5 <sup>×</sup> 102 ICC/μL, 86 pmol/min and 23.4% HNAP in CI, and to 4.1 <sup>×</sup> 102 ICC/μL and 21.6% HNAP in D2 due to process-related oxygen and external biomass entries and a rainfall event (CI).

Ozonation eliminated most intact organisms to 2.0 <sup>×</sup> 101 ICC/μL and 25 pmol/min in CI based on its high redox potential, but at the same time this process provides organic substrate: dead cell material to still intact organisms. This conjuncture is also confirmed when comparing the lysed cell gates in Section IV and Section V in Figure <sup>4</sup> with 2.3 <sup>×</sup> <sup>10</sup><sup>2</sup> desolate cell count (DCC)/μL (aeration) and 2.0 <sup>×</sup> <sup>10</sup><sup>2</sup> DCC/μL (ozonation). Also, the wide spread of standard deviation results, regarding HNAP, indicates that the amount of present cells is below detection limit since standard deviation values surpass the mean value by 40.6%. As a result, biomass increased tenfold in the sandfilter effluent in CI. This relation however, could not be observed with the BACTcontrol method where ALP-TEA slightly decreased to 23 pmol/min. BRU in D1 and D2 decreased by 46% (D1) and 51.6% (D2) indicating that sandfilters act as a microbiological barrier.

**Figure 3.** Comparison of FCM intact cell count in (**a**) D2, (**b**) D1 and (**c**) CI, (ICC, n = 5) with total enzymatic activity (ALP-TEA, n = 3 only CI), Sampling Dates: 9 May 2019 (CI), 15 May 2019 (D2) and 5 June 2019 (D1), FCM samples were stained with SYBER® Green and propidiumiodide, error bars representing standard deviation for those points.

Also, no significant impact of UV disinfection in CI on ICC as well as HNAP and LNAP amount was observed. ICC decreased by 16% to 2.0 <sup>×</sup> <sup>10</sup><sup>2</sup> ICC/μL and HNAP by 0.33%. UV-C treatment only causes damages of the bacterial genome but has no impact on bacterial cell membranes which is also confirmed in Section II and Section III in Figure 4 [22]. GAC filter effluents show an increase of BRU by 63.5% (D1) and 56.4% (D2). For GAC filter effluents in D1, BRU increase may be explained by an additional aeration process after water extraction and/or biological active GAC filters. The cell amount of 2.7 <sup>×</sup> 102 ICC/μL in D2 with a decrease of HNAP by 56.4% at the same time in the GAC filter effluent may indicate that certain strains, especially HNAC, are more adept at colonizing the filter surface and may be able to out-compete LNA cells [23].

Final disinfection using chlorine leads to an ICC decrease by 95.0% to 1.0 <sup>×</sup> 101 ICC/μL (CI), by 92.1% to 4.0 <sup>×</sup> 10<sup>1</sup> ICC/μL (D1) and by 88.6% to 3.0 <sup>×</sup> 10<sup>1</sup> ICC/μL (D2). HNAP decreased by 90.1% to 7.2% (CI), by 50.8% to 13.3% (D1) and by 25.4% to 5.0% (D2). ALP-TEA was detected in CI with 18 pmol/min which is in accordance with other studies [16,22]. Interestingly, the HNAP portion in

D2 decreased by 85.2% and the LNAP amount increased correspondingly. This may indicate that ClO2 damages HNAC membranes faster and more effectively than LNAC membranes which was also observed in several water samples from a previous study [22]. Alternatively, there may be a correlation between perception-events and rising biomass in GAC effluents.

**Figure 4.** Dot-plots of water treatment trains in CI, Line 1: RBF, aeration, ozonation; Line 2: sandfilter, ultraviolet (UV-C) disinfection, assignments of Sections I to III: Background Signal, Intact LNAC, Intact HNAC, Sections IV-V: Lysed Cells, x-axis: Flourescence Signal 1 (FL1), y-axis: Flourescence Signal 2 (FL2).

The results shown in Figure 5 confirm the correlation between the attenuation of microorganisms and travel time in the aquifer. ICC decreased by 1.10 log units to 1.5 <sup>×</sup> 10<sup>2</sup> ICC/μL in well BF1 while ICC in well BF4 with a travel time of 100 ... 220 days decreased by 1.61 log units to 4.7 <sup>×</sup> 10<sup>2</sup> ICC/μL which is in accordance with the D2 results of average retention of 1.22 log units (not all sampling days are shown in Figure 5). The limit of detection using the HPC method is also demonstrated in Figure 5. While 220 MPN/mL and 48 MPN/mL were determined in Danube River water, no colonies could be detected in water from the RBF wells. These results suggest that new rapid microbiological methods (e.g., FCM) could be powerful tools for monitoring general microbiological water quality during treatment and distribution, as well as for the design and optimization of RBF site operations.

**Figure 5.** ICC monitored in various RBF wells in Budapest and their travel times in days, BF 1 = Csepel water plant, well No. 1, BF 2 = Tahi 1, well No. 5, BF 3 = Tahi 2 well No. 5, BF 4 = Szigetujfalu, well No. 7 with error bars rep-resenting standard deviation for those points, FCM samples were stained with SYBER®Green and propidiumiodide, HPC data were provided by Budapest Water Works Ltd. Modelled travel times may overlap depending on Danube River water levels.

#### *3.2. Ultrafiltration Pilot Plant and Online FCM Measurements*

During operation of the ultrafiltration pilot plant, bacteria and other particulate matter were efficiently retained independent of feed-water quality. Figure 6 shows online FCM measurement results of flocculated river water and river water as feed water. TCC in flocculated river water was fluctuating between 3.8 <sup>×</sup> <sup>10</sup><sup>3</sup> and 1.2 <sup>×</sup> 103 TCC/μL. The 30 cycles from 9 November to 12 November showed overall stable TCC of 2.1 <sup>×</sup> 103 TCC/μL but also daily fluctuations, and two biomass peaks (cycle 9 and cycle 29).

TCC was considerably higher in Elbe River water with an average amount of 9.4 <sup>×</sup> 10<sup>3</sup> TCC/μL with daily fluctuations from 4.4 <sup>×</sup> 104 to 1.9 <sup>×</sup> 103 TCC/μL (Figure 7) and biomass peaks at 27 April (cycle 12) and 3 May (cycle 11) caused by a rainfall event on April 26th, and due to several rainfall events around 1 May and 4 May. LNAP increases during rainfall and decreases during dry periods by 74.5% (not shown in Figure 6) as well as the TCC amount which was also documented in two studies of Besmer et al. [24,25]. TCC in Permeate 1 range from 2.9 <sup>×</sup> 103 to 4.0 <sup>×</sup> <sup>10</sup><sup>1</sup> TCC/μL, and in Permeate 2 from 2.6 <sup>×</sup> 103 to 3.7 <sup>×</sup> <sup>10</sup><sup>0</sup> TCC/μL dependent on the feed water quality in the corresponding cycle but are in a normal range due to inevitable bacterial regrowth after treatment. Cell numbers in the range from 101 to 10<sup>2</sup> cells/μL of diverse bacterial dynamics in similar water habitats such as riverbank filtrate or spring water were reported to be normal in previous studies [13,24,26]. However, the authors are not aware of any online FCM long-term studies that provide data of a similar ultrafiltration pilot setup.

With respect to the influence of feed water quality no significant difference in permeate quality using river water or flocculated river water as Feed (*t*-test, n = 44 (Elbe River), n = 59 (flocculated Elbe River water) *p* > 0.05) was reported, demonstrating the high performance of ultrafiltration in terms of microbiological removal of bacteria. Moreover, all online FCM results revealed higher amounts of TCC (3.6 fold) in Permeate 1 in comparison to Permeate 2 due to the fact that the membrane units in this pilot study differ in membrane area per module and pore size [19]. Hence, unit 1 with a pore size of 20 nm and 60 m2 is more permeable to microorganisms than unit 2 with 18 nm and 55.7 m2.

**Figure 6.** Continuous determination of total cell count (TCC) in flocculated Elbe River water (feed), Permeate 1 and Permeate 2, (**a**) 9–12 November 2018, (**b**) 19–22 November 2018), samples were stained with SYBER® Green.

**Figure 7.** Continuous determination of total cell count (TCC) in Elbe River water (feed), Permeate 1 and Permeate 2, (**a**) 26 April–29 May 2019, (**b**) 2 May–4 May 2019, samples were stained with SYBER®Green and propidiumiodide.

When investigating the cut-off values of Permeate 1 and Permeate 2 (Figure 8) median microbiological removal rates of 2.41 and 1.19 log units were observed in flocculated river water feed and Elbe River water feed in Permeate 2, whereas log removal efficiencies of only 1.28 and 0.92 log units were achieved in Permeate 1. Such differences were also reported by Haas et al., [19] regarding ATPt results at the same pilot setup. Log units in Elbe River water were in addition 28.1% (Permeate 1) and 50.6% (Permeate 2) lower than in flocculated river water due to the membrane flux being influenced stronger by particular matter (e.g., dissolved iron or manganese). This leads to severe fouling problems, particularly biofouling and organic fouling on the membrane surface [27].

**Figure 8.** Differences in log removal rates of Permeate 1 and Permeate 2 in (**a**) flocculated Elbe river water *t*-test, n = 59, *p* < 0.05, n = 59 and (**b**) Elbe River water, t-test, n = 44, *p* < 0.05.

#### **4. Discussion**

In this study, the applicability of FCM and two enzymatic detection methods (ATP, ALP-TEA) for monitoring water-quality parameters at RBF sites was investigated. Despite a good correlation between FCM and ALP-TEA values, several differences were observed. In general ALP-TEA values correspond better to FCM results. This may be due to the fact that ATPi amount varies across living organisms and species, and is inter alia dependent of physiological states, especially HNA cells contain a higher ATPi amount than LNA cells [28]. Furthermore, ATPi is calculated from ATPt and ATPe results which were measured with a hand-held device where no collection of a fixed sample volume was possible. Therefore, values were fluctuating which is proven by the standard deviation on the triplicate samples.

The results of the dilution series (Figure 2) and also ALP-TEA values in CI (Figure 3) differentiate from the ICC trend as well as from the ATPt results. This could be caused by the nature of intracellular enzymatic methods. Intracellular enzymatic activity is generally bound to enzymatic concentration which is mostly dependent on bacterial state of growth while no information about species, size or number of a general population pattern is obtained.

ALP-TEA activity is especially high in the exponential stage of growth [13]. The probability of exponential growth in smaller bacterial communities is lower due to its limited number of microbiological species which is confirmed in Figure 2. Besides concentration, enzymatic processes are also dependent on specific reaction conditions which can be disturbed by interfering substances such as iron and manganese compounds. This was observed during ATP measurements in RBF well water samples (not shown in Figure 5) which usually contain higher concentrations of dissolved iron and manganese.

Additionally, FCM results may provide false positive signals due to staining limitations. SYBER®Green binds to any source of DNA including higher animals and plants. Although dividing cells and background signals through gates will remove most errors in quantitation, an overestimation

of particle numbers must be always taken into account when assessing FCM results. Moreover, a recent study revealed that propidiumiodide-based viability staining can significantly overestimate DCC due to the presence of extracellular nucleoid acid (eNA) biofilms [29]. False positive results with propidiumiodide dye have been also associated with high membrane potential or might be influenced on physiological processes other than membrane damage in earlier studies [30,31].

Furthermore, the removal of microorganisms was correlating with travel time, proven by FCM and HPC measurements for various RBF wells in Budapest. The observed log removal rates for bacteria during RBF were in agreement with average values of 1.5 ... 3.5 [32]. Despite the short travel time of 2 ... 5 d in BF1, the removal rate of 1.10 log units is only slightly lower than values found at other RBF sites [14,15]. No colony counts were detected by HPC or ALP-TEA (not shown in Figure 3), indicating the limitations of these methods, whereas by FCM and ATP methods a low limit of detection was proven and a high efficiency to assess microbiological dynamics during RBF.

Changes in BRU concentration after aeration processes were observed in D1 and CI, both showing a slight increase in cell concentration by 4% and 13%. Here the aeration is operated as open cascade towers and microbiological growth is stimulated by external biomass entries, especially seasonal pollination. Further research is needed since analyzed data of the aeration processes suggest a connection between bacterial growth and pollination. This may be based upon the assumption that, especially on days without rainfall events, the growth was more intense. Alternatively, cell growth in aeration towers D1 and CI could be caused by FCM dyes on pollen entries. This may lead as previously mentioned to an over estimation of fluorescence signals.

Furthermore, differences in microbiological removal efficiency during sand filtration were observed at all sites. Previous ozonation provides dead cell material, a carbon source, to still intact microorganisms which are able to pass the sandfilter, and are therefore responsible to BRU increase in the filter effluent. In D1 and D2 however, sand filtration operated as a microbiological barrier dependent on cell size, cell morphology, motility, and membrane surface chemistry [33]. Additionally, microbiological removal efficiency is dependent on the sandfilters general conditions such as the thickness of the uppermost biofilm (*Schmutzdecke*), sand composition, or maturation of microbiological community in the *Schmutzdecke* [34].

In terms of BRU increase in the D2 GAC filter effluent and the corresponding HNAC decrease, it is presumed that certain strains, especially HNAC adapt more at colonizing the filter surface. In a pilot study by Shirey et al. [23], the community structure of heterotrophic bacteria associated with three GAC and two anthracite filters was examined over 12 months. Besides the diversity of bacterial community structures in prefiltered water, media composition, and depth, time was also a significant factor influencing bacterial communities within the filters. After the initial acclimation and colonization period had passed, overall community structure became less variable [23]. To clarify whether those cells are composed of a high HNAP amount, a similar pilot study using online FCM is necessary to prove this assumption.

Final disinfection results in a strong BRU decrease which was observed at all sites. Interestingly, in comparison to CI and D1 a decrease by 25.5% HNAP was observed in D2. This was likely due to the fact that ClO2 was used as a disinfectant instead of Cl2. Previous studies revealed cell membranes of HNAC bacteria were damaged much faster than those of LNAC bacteria during treatment with ClO2 while only small differences were observed during treatment with chlorine and chloramine, and no difference was observed for ferrate treatment [22]. Thus, the considerably high HNAP decrease of 85.2% in D2 may be due to ClO2 treatment.

Results from the ultrafiltration pilot plant study confirmed that online sensors are of advantage in terms of microbiological monitoring. Online measurements allow real-time detection of stable phases which are missed by grab sampling or incorrectly characterized. Optionally, online FCM can be coupled with online enzymatic activity probes (e.g., BACTcontrol). Besides ALP-TEA enzymatic activities of EC (β*-galactosidase*) and *E. coli* (β*-glucoronidase*) can be determined that may speed-up the detectability of incidents which impair microbiological water quality and safety [16]. More detailed evaluation regarding advantages of microbiological online sensors is given in previous studies [16,24–26].

Ultrafiltration proved to operate as an efficient barrier against microorganisms. During online FCM measurements, the average TCC values measured in river water as well as in flocculated river water are in normal range of surface water [35]. All feeds indicate an apparent bacterial fluctuation as well as one event linked to rainfall (cycle 13 on 27 April). According to a local meteorological station in D1, a rainfall event of 16.3 mm was recorded on 26 April in the evening [36]. This event is in accordance with the observed TCC peaks. The second event indicates a strong TCC increase of 65.9% (cycle 11, 3 May) which is likely caused by several entries of different origin (e.g., shipping or leisure activities) into the Elbe River due to a national holiday event on 1 May, and minor rainfall events including 3 May (0.2 mm) [36].

To underline those assumptions, abiotic parameters such as EC, pH, or dissolved oxygen should have been measured directly in feed waters which was beyond being integrated into the experimental setup. Diurnal fluctuations of abiotic parameters in river water have been investigated in previous studies [24,37,38]. Microbiological changes in surface waters may therefore be linked to photosynthetic and respiratory metabolic processes as well as to changes in precipitation and dissolution of ions which was inter alia assumed by Besmer et al. 2014 [24].

While the authors are not aware of previous studies regarding online FCM during ultrafiltration, the TCC amount detected in permeates are in accordance to typical values of riverbank filtrate or spring water. The daily patterns in Figures 6 and 7 were in accordance with microbiological dynamics in feed water. The occurrence of TCC in the permeates was probably due to bacterial regrowth after treatment [26,27].

Despite general membrane characteristics such as pore size and membrane surface and thus differences in permeate quality were observed (Figure 8), median removal rates in river water (0.92 log units Permeate 1 and 1.19 log units Permeate 2) was less than in flocculated river water (1.28 log units Permeate 1 and 2.41 log units Permeate 2). Jadoun et al. 2018 suggested that biofouling promotes the establishment of populations of water-borne pathogens on membrane surfaces. This was proven in monitoring studies using both culture-based and quantitative polymerase chain reaction (qPCR) methods in which the ability of microorganisms to establish rapid biofilm formation and persist on the membrane surfaces was demonstrated [27].

Considering the average amount of TCC in Elbe River water which is five times higher than in flocculated river water, the impact of biofouling on the membrane surface would most likely affect microbiological removal efficiency. The survival of biofilm-forming bacteria on the membrane surface despite in situ backwash and chemical treatment highlights the importance of optimization of the ultrafiltration process e.g. by applying RBF as a pretreatment step to ultrafiltration. In addition, if coupling RBF and ultrafiltration, fewer chemicals would be needed with regards to disinfection and membrane backwashing.

#### **5. Conclusions**

The results in this study provide insights into microbiological quality of treated water and emphasize the value of rapid, easy, and sensitive alternatives to HPC-based monitoring techniques. Despite a good parameter correlation, in particular in between ICC and ALP-TEA results, several divergences were observed. These were probably caused by method-based limitations such as staining techniques (FCM) or interfering ions (ATP, ALP-TEA). Also, due to different parameters defining bacterial concentration being measured, a direct comparison between these variables was rather difficult. The sets of microbiological data of RBF and other water treatment-related samples, nevertheless, enhance the understanding and improve the assessment of microbiological dynamics during drinking water treatment. However, further research, especially long-term studies to cover different seasons at RBF sites, are of vital importance to underline the results gathered in this study.

The online FCM data during ultrafiltration revealed diurnal BRU fluctuations, likely in response to nutrient concentration and abiotic parameters in feed water. These rapid changes should be considered when water is monitored by grab sampling. At present, little is known about online FCM during UF. Further long-term data combined with abiotic parameter monitoring and optional enzymatic activity are, therefore, required.

The presented methods may serve as possible alternatives in the future for the assessment of the quality of RBF water. However these methods are not yet accredited, hence measurement results are informative. Nonetheless, the article is of interest to all drinking-water suppliers that operate RBF systems to consider improving general microbiological monitoring.

**Author Contributions:** Y.A. reviewed previous literature and prepared the article draft, T.G. initiated the research and acquired the funding, Y.A., G.-H.O. and M.P. carried out measurements, R.H., R.B. and Z.Á.N.-K. organized sampling campaigns and operation of treatment facilities, J.A. introduced ALP-TEA measurements and provided the device, and all authors reviewed the final manuscript. All authors have read and agreed to the published version of the manuscript.

**Funding:** All primary data was collected within the AquaNES project. This project has received funding from the European Union´s Horizon 2020 Research and Innovation Program under grant no. 689450.

**Acknowledgments:** The authors gratefully acknowledge support from the DREWAG NETZ GmbH, Budapest Waterworks Ltd, Hungary, microLAN B. V.

**Conflicts of Interest:** The authors declare no conflict of interest. The founding sponsors had no role in the design of the study; in the collection, analysis or interpretation of data; in the writing of the manuscript; or in the decision to publish results.

#### **Abbreviations**

The following abbreviations are used in this manuscript:


#### **References**


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

MDPI St. Alban-Anlage 66 4052 Basel Switzerland Tel. +41 61 683 77 34 Fax +41 61 302 89 18 www.mdpi.com

*Water* Editorial Office E-mail: water@mdpi.com www.mdpi.com/journal/water

MDPI St. Alban-Anlage 66 4052 Basel Switzerland

Tel: +41 61 683 77 34 Fax: +41 61 302 89 18