Desalination of saline farmland drainage water through wetland plants

Agricultural Water Management 156 (2015) 19–29

Contents lists available at ScienceDirect

Agricultural Water Management journal homepage: www.elsevier.com/locate/agwat

Desalination of saline farmland drainage water through wetland plants Y.N. Yang a,∗ , Q. Sheng a , L. Zhang a , H.Q. Kang a , Y. Liu b a

School of Chemistry and Environment, Beihang University, 37rd Xueyuan Rd, Haidian District, P.O. Box 106, 100191 Beijing, PR China Department of Water Resources, China Institute of Water Resources and Hydropower Research, 1, Yuyuantan South Rd, Haidian District, 100038 Beijing, PR China b

a r t i c l e

i n f o

Article history: Received 29 May 2014 Accepted 2 March 2015 Keywords: Saline water discharge Salt removal Typha spp. Phragmites communis Potamogeton crispus

a b s t r a c t To protect against soil secondary salinization, a desalination process for farmland drainage using wetlands was evaluated. In this study, the desalination effects of different plants in Chagan Lake were analyzed. A field experiment was conducted in the Qianguo irrigation district to choose the most efficient desalting plant by comparing the biomass contents and the ash rates of Typha spp., Phragmites communis, Phragmites japonica Steud. var. prostrata (Makino) L. Liu, Medicago sativa Linn., Lemna minor L. and Potamogeton crispus. Typha spp., Phragmites communis and Potamogeton crispus performed best among tested species in removing salt from saline farmland drainage. According to the calculated ash rates and ion contents, the amount of salt removed by reaping reed and cattail accounted for 10–26% of the salt in the drainage. The removal efficiencies of Ca2+ , Mg2+ , Na+ , Cl− and SO4 2− ions are 9–15% per year. A constructed wetland containing 233–288 km2 of Typha spp. is required so that the removal efficiency of these six ions can be more than 80%. © 2015 Published by Elsevier B.V.

1. Introduction Saline land, as a kind of land resource, is being paid more and more attention to its improvement and utilization these days. It is estimated that there are 9.5 × 108 ha saline area worldwide, accounting for 7.26% of the total land area on earth (Malcolm and Sumner, 1998). In China, the total saline area is 1.0 × 108 ha (Xiong, 2005), accounting for about 10% of that of the world, while 80% of which has not been modified and developed (Song, 2009). Songnen Plain, with 3.2 × 106 ha saline area which accounts for 19% of the total plain area, is one of the five salinization land distribution areas in China (Li, 2000). Salinity in Western Songnen Plain experienced a complex geological transformation, related to the factors of stratum, hydrogeological conditions, groundwater runoff and surface runoff, tectonic framework, etc. Salinization of Qianguo and Da’an areas were attribute to historical reasons of Da’an ancient river channel – the role and processes of its tectonic movements provides salt sources and a means of transport for the formation of salinity and to create salt accumulation environment. Songnen Plain, which soil salinity was reduced by going through irrigation leaching, achieved the goal of graining production of

∗ Corresponding author. E-mail address: [email protected] (Y.N. Yang). http://dx.doi.org/10.1016/j.agwat.2015.03.001 0378-3774/© 2015 Published by Elsevier B.V.

15 billion kilograms in Northeast China (Yang and Liang, 2007). The total construction area of changing saline to paddy was 3.41 × 105 ha. Rice cultivation can be used to treat the saline land in the Songnen Plains, opening up the area affected by salinity for land development and leading to construction around the irrigation area in the western part of the northeast. However, surveys found that because of the lack of supporting evaluation for drainage water treatment, the problem of soil secondary salinization increased. The secondary salinization turned non-saline–alkali soil into saline–alkali soil, and the crop yield decreased (Yang et al., 2009). Therefore, problem in the treatment of saline farmland drainage water must be solved for the vicious circle of regional agricultural development and ecological environment improvement. As a result, food production can be expanded while the surrounding environment is well protected (Yang and Liang, 2007; Li et al., 2009). To avoid the secondary salinization of soil, this project proposed to use wetland as receiving area of saline water. On the one hand, retreated saline water was used to irrigate the wetlands; on the other hand, the treated discharge water of the wetlands can be recycled in the saline area to keep a sustainable use of saline water. Chagan Lake, as broad as 480 km2 , covers two counties and one city in the west of Jilin province: Qianguo County, Qian’an County and Da’an city (Li, 2013). Chagan Lake is a national nature reserve and also a saline wetland recorded in the list of China’s wetland. The whole lake is 37 km long from south to north, 17 km wide from

20

Y.N. Yang et al. / Agricultural Water Management 156 (2015) 19–29

east to west, with meandering Lake Shoreline 128 km long, known for fishery production, agricultural production and natural resort (Sun, 2014). The south part of the Chagan Lake, which is considered as the core area of the Chagan Lake reserves, is about 3.33 × 103 ha and has a water capacity of 7 × 108 m3 (Zhang et al., 2013). It is a main bird habitat. Because of its biological abundance and various aquatic plants such as Typha spp., Phragmites communis and the Potamogeton crispus, etc., the south of Chagan Lake is the return water intake area of the Qianguo irrigated regions named as the paddy field of saline–alkali soil in Songnen plain. About 1.3 × 108 m3 –2 × 108 m3 of drainage water from the saline–alkali soil in the Qianguo irrigated region flows into the south of Chagan Lake every year. Qianguo irrigated region, which the Yinsong channel flows by, has an area of 1285.7 km2 and 41.7% of the total area is salinization land (Qin et al., 2002). There is a controversy on whether to discharge recycled water from the saline–alkali paddy field into the Chagan Lake wetland reserve. The consenters think that the drainage water from saline–alkali soil could provide water to the Chagan Lake wetland, which is beneficial to the wetland ecosystem restoration; while the oppositions think that recession salinity from the saline–alkali paddy field is so high that it can cause adverse effects to the Chagan Lake wetland ecosystem. Therefore, a better choice is to select the south of Chagan Lake carrying large amounts of drainage water from the saline–alkali soil as the research object. At present, most of the salinity wastewater treatment methods are physical and chemical treatments, including electro dialysis, reverse osmosis, distillation and ion exchange, etc. (Lu et al., 2005; Fan, 2003; Darwisha et al., 2009). Yao (2004) has used a membrane integrating micro filtration, ultrafiltration and reverse osmosis to study the treatment of high-salinity wastewater. The research indicated that the desalination rate could reach 96.5% when the salt content of the wastewater was 2000–5000 mg L−1 . The desalination rate of electro dialysis could be as high as 99.3%, but the cost was relatively higher. To remove salt from the saline farmland drainage, Lee et al. (2003) have studied the saline farmland drainage treatment using a reverse osmosis membrane. The research showed that the treatment efficiency of the ESPA-1 and NF-90 membrane was higher with a desalination rate of 95% when the concentration of sodium was 1150 mg L−1 . In Egypt, desalting processes are widely used for farmland drainage treatment, including electro dialysis, reverse osmosis and ultrafiltration. Abulnour et al. (2002) has demonstrated that the biological ultrafiltration membrane system is the cheapest method by comparing the economy of various different methods. The system uses microfiltration/ultrafiltration scheme in the pretreatment, then followed by reverse osmosis. The saline water from the reverse osmosis is further treated to increase system recovery. However, these techniques are still problematic due to complex operation and high cost. Membrane treatment has problems with membrane fouling and concentrated water treatment. The high salt concentration and complex ion composition in the concentrated water result in unsatisfactory treatment efficiency (EI-Zanati and EI-Khatib, 2007; Gilron et al., 2000). In general, the best way to improve the saline–alkali land is to grow salt-tolerant plants, because halophytes have good absorption of salt. In a saline environment, halophytes can absorb salt from their growing environment to reduce the intracellular water potential for remitting salt stress so that the trans-membrane transport of water flows in a favorable direction for cell growth. For example, Zhang (2005) studied the cultivation of salt tolerant species including sainfoin, Medicago sativa Linn., Symphytum officinale, Coronilla varia, and Festuca arundinacea Schreb, on saline farmland. The results show that halophytes have an obvious desalination effect, and the desalination rate of halophytes can reach 31.1% and 19.1% in a saline soil layer of 0–20 cm and 0–100 cm,

respectively. Similarly, this paper presents a study that uses haloduric hydrophytes to absorb salt and remove salinity of drainage water. Halophytes have an aptitude for salt absorption (Zhao et al., 2002; Ren et al., 2004). Lymbery et al. (2006) from Australia has studied the use of Juncus sp. to address the saline–alkali farmland, and the results showed that the desalination rate can reach as high as 54.8%. Yang et al. (2004) used the mode of fish-rice-reed-cattail to improve soda saline–alkaline land. The results showed that the desalination effects of cattail and reed were much better. The purpose of these studies was to investigate salt removal in saline–alkali soil using halophytes. Therefore, using halophilous hydrophytes to remove the salt from farmland drainage water may be a better choice. In order to find a new way to manage the drainage water from saline–alkali land, this paper analyzed the desalination effect of different aquatic plants in Chagan Lake. A field experiment was conducted in the Qianguo irrigation district to choose the most efficient desalting plant by comparing the biomasses and the ash rates of Typha spp., Phragmites communis, Phragmites japonica Steud. var. prostrata (Makino) L. Liu, M. sativa Linn., Lemna minor L. and Potamogeton crispus.

2. Materials and methods 2.1. Description of the experimental site The study area is located at Chagan Lake near the Da’an irrigation district. Chagan Lake is the main receiving area for the saline farmland drainage water from the Qianguo irrigation area, and the annual yield of drainage water from the Songhuajiang channel flowing into the Chagan Lake is approximately 1.3 × 108 –2 × 108 m3 . The area of Chagan Lake is large, and the water quality and quantity are stable. The climate of Chagan Lake is semi-arid sub-humid continental monsoon. There are 2906 h of sunshine in Chagan Lake area all year round, and the annual solar radiation is 517.5 kJ/cm2 . So this region has abundant light resource. The annual average air temperature of the Chagan Lake area is 5.0 ◦ C, with a lowest temperature of −30.9 ◦ C and a highest temperature of 33.9 ◦ C. The frost-free periods are 141 days and the average evaporation values for the experimental area is 1466 mm per year. The total area of the Qianguo irrigation district is 1285.7 km2 and the salinization of land accounts for 41.7% of the total irrigation area. Generally speaking, the areas of the saline soil on both sides of the Songhua River channel (paddies drainage main channel in Fig. 1) are large, and the salinity-alkalinity of the irrigation district drainage water is high. Chagan Lake, the main receiving area for the saline farmland drainage water from the Qianguo irrigation area, has diverse species of biomes. Many aquatic plants grow in Chagan Lake, such as Typha spp. and Phragmites communis. Furthermore, the difference in the salt content between southern part and northern part of the lake is large enough to study the desalination of halophytes growing in different environments. Experimental zone geographic locations are shown in Fig. 1.

2.2. Methodology According to the recession mechanism of rice paddies in saline–alkali land and to ensure the continuity and periodicity of the sampling, the sampling frequency of the water and soil was set as follows: twice in the wet period (late July and early September), twice in the level water period (middle of May and middle of June) and once in the drought period (middle of November). According to the plant growth cycle, the plant sampling was set in the level

Y.N. Yang et al. / Agricultural Water Management 156 (2015) 19–29

21

Fig. 1. Experimental zone geographic locations.

water period (middle of May) and at the end of wet period (late August). 2.2.1. Sample collection a. Water and soil sample collecting points Water and soil sampling points were set respectively at drainage channel, South Lake Wetland, South Lake Centre and North Lake entrance. Water samples were sampled at 0.2–0.4 m below the surface, and soil samples were sampled at 10 cm below the surface. b. Plant sample collecting points Twenty-two sampling points were set in South Lake wetland of Chagan Lake, while 5 sampling points were set in North Lake wetland. The reason why we set more sampling points in the South Lake than the North Lake is that more diverse of plants grow in South Lake wetland. Moreover, the drainage flows into the South Lake and is treated by plants in South Lake first, then flows into the North Lake. Typha spp., Phragmites communis, Phragmites japonica Steud. var. prostrata (Makino) L. Liu, M. sativa Linn., L. minor L. and Potamogeton crispus samples were collected at each sampling point. According to the Fresh Water Biological Resources Survey Technical Specification (DB43/T 432-2009), for the emerging plants, such as Typha spp., Phragmites communis, Phragmites japonica Steud. var. prostrata (Makino) L. Liu, M. sativa Linn., each sampling point was 1 × 1 m2 in size, while for the floating-leaved plants and submerged plants, such as the L. minor and Potamogeton crispus, samples were using the aquatic plant clips which sampling area is 0.25 m2 . 2.2.2. Sample preparation a. Water sample preparation Specific sampling methods were referred to “water quality sampling program design requirements (HJ495-2009)”. This standard specifies items of samples including a variety of water, sludge and sediment at the bottom of the water, and guidelines

for quality control, quality characterization, sampling technical requirements and the principle of pollutants identification sampling scheme. Sampling water was taken by putting sampling bottles into the water, filtering the impurities floating on the water surface and underwater sediment with a strainer. Then the samples were stored after recording the sampling time and depth. When collected water samples cannot be analyzed immediately and need to be transported to the laboratory, “water sample preservation and management provisions (HJ493-2009)” specifies the method to preserve water samples by adding protective agent to the sample. It also specifies samples transportation, acceptance and other details to ensure the quality of stored sample. The sampling volume taken with the plastic bottle at each sampling point was 500 mL. A few drops of sulphuric acid (1 + 3) was added as fixative to prevent microbial decomposition of organisms and ensure the accuracy of the measurement. The mixture was then kept at 2–5 ◦ C in a refrigerator and processed for the physico-chemical parameter measurements as soon as possible. b. Soil sample pre-treatment Soil samples were collected using artificial drilling at a depth of 10 cm. Every soil sample of approximately 500 g was put into the sampling bag and stored. Soil sample process included drying, grinding, sifting, mixing, bottling. Specific steps were as follows. The soil samples were placed on wooden or plastic plate, spread into a thin layer and placed indoor with ventilation to dry. When soil samples were half-dry, big clods were crushed into pieces. Air-dried samples were separated from plants and animal residues and clods. Drying time was about 2–3 days. The dried soil samples were grinded with a wooden pestle to make sure they could be sieved through 2 mm sieves. After mixing, the samples were divided into two parts by quartation for physical and chemical analysis, respectively. The first step of quartation was to mix and collect the soil samples, then pave

22

Y.N. Yang et al. / Agricultural Water Management 156 (2015) 19–29

Table 1 The determination method of water and soil indicators. Monitoring indicators

Monitoring methods

Temperature pH Conductivity COD

Thermometer pH meter Portable conductivity meter Rapid digestion spectrophotometry

NH4 +

N

Acid spectrophotometric method

−

N N

Ion chromatography Ion chromatography Alkaline potassium persulfate digestion-UV spectrophotometric method Molybdate spectrophotometry

NO3 NO2 − TN TP

CO3 2− and HCO3 − −

2−

Cl , SO4 Na+ , K+ , Ca2+ , Mg2+

Titration Ion chromatography Atomic absorption spectrophotometry

them into a square with four parts divided diagonally. Two parts of diagonal soil were kept for further use. If the weight of the soil retained was more than needed, then the soil was quartered again until the weight of the two diagonal parts met the requirement. The soil samples for chemical analysis were grinded and sieved through the 1 mm aperture sieve. The samples were preserved with plastic bottles for six months or a year to prepare for checking purposes when necessary. c. Plant sample preparation According to the growing area and density of the plants, different kinds of plants were collected in the South Lake and North Lake. In North Lake, three sampling points were selected to collect Typha spp. and Phragmites communis, two sampling points were for Phragmites japonica Steud. var. prostrata (Makino) L. Liu and M. sativa Linn., while in South Lake, fourteen sampling points were selected to collect Typha spp. and Phragmites communis, eight points were for L. minor Linn. and Potamogeton crispus. According to the different local plant harvesting methods used and the growing characteristics of different plants, Typha spp. and Phragmites communis are the emergent aquatic plant, so 10–20 cm above the ground were harvested. On the other hand, though Phragmites japonica Steud. var. prostrata (Makino) L. Liu is emergent aquatic plant as well, it grows in the dry land and it is shorter than the Typha spp. and Phragmites communis, so we cut it 5 cm above the ground. M. sativa Linn. was harvested along the ground. L. minor L. is a floating plant, and Potamogeton crispus is a submerged plant, so they were sampled by dredge. The stems and leaves of Typha spp., Phragmites communis, Phragmites japonica Steud. var. prostrata (Makino) L. Liu and M. sativa Linn. were cut and measured to 5–10 cm, and were gathered in groups of 500 g. The plant samples were dried for 5 h at 50 ◦ C and were crushed and mixed uniformly. They were then heated to 105 ◦ C for 5 h. A 1 mm sieve was used to filter the samples before determining the ash rate and ion contents (Bao, 2005). 2.2.3. Samples analyses a. Water and soil samples analyses The methods of water and soil analyses are shown in Table 1. b. Plant samples analyses (1) 10 strains of the same species were randomly selected in each sampling point and each strain weight was measured with a spring balance to determine the biomass of each species. A meter ruler was used to measure the height of the strain.

Methods source

HJ/T 399-2007: water quality-Determination of the chemical oxygen demand-Fast digestion-Spectrophotometric method HJ 536-2009: water quality-Determination of ammonia nitrogen-Salicylic acid spectrophotometry

GB/T 11894-89: water quality-Determination of total nitrogen-Alkaline potassium persulfate digestion-UV spectro photometric method GB/T 11893-89: water quality-Determination of total phosphorus-Ammonium molybdate spectrophotometric method DZ/T 0064.49-93: underground water quality testing method-Titration determination of carbonate ions and bicarbonate root and hydroxyl GB 11907-89: water quality-Determination of silver-Flame atomic absorption spectrophotometric method

(2) The number of strains per square meter was calculated in each sampling point and the average number of strains was the plant growth density. (3) The moisture content and ash rate were determined according to the book “Agrochemical Analysis of the Soil” (Bao, 2005). A clean aluminum box was dried in a 105 ◦ C thermotank for 30 min, then cooled down to ambient temperature in a desiccator and the weight m0 recorded (accurate to 0.001 g); 3–5 g grinded and well mixed wetland hydrophyte sample was added into the aluminum box to accurately measure the total weight of the aluminum box and the sample (m1 ) before oven drying. The aluminum box was dried at 50 ◦ C for 5 h and then at 105 ◦ C for another 5 h, cooled down to ambient temperature in a desiccator (about 20 min) and the weight m2 recorded immediately. The moisture content was calculated with the following equation: Moisture content =

[(m1 − m0 ) − (m2 − m0 )] (m1 − m0 )

m0 – the weight of dried empty aluminum box m1 – the weight of aluminum and sample before oven drying m2 – the weight of aluminum and sample after oven drying A clean and empty melting pot was weighed as m0 ; 5 g dried and crushed wetlands aquatic plant samples were added into the empty melting pot and heated slowly till the samples carbonized. Then the melting pot was heated in a muffle furnace at 550 ± 25 ◦ C for 6 h, cooled down to ambient temperature in the desiccator and the total weight of the melting pot and ash m1 recorded. The ash was washed into a 100 mL beaker from the melting pot with 25 mL hot distilled water, filtered with the ashless filter paper by adding more water till the total volume of the solution reached 150 mL. The filter paper and residue were put back into the melting pot and heated in a boiling water bath. Then the melting pot was transferred into the muffle and burned till no char under 525 ± 25 ◦ C (about 1 h), cooled down to ambient temperature and the weigh recorded. Then the melting pot was put back to the muffle and burned for 30 min, cooled down, and the weight measured. The operation was repeated till the difference between two measurements was no more than 0.001 g. The minimum weight was chosen as total weight of the melting pot and water-insoluble ash m2 .

Y.N. Yang et al. / Agricultural Water Management 156 (2015) 19–29

23

Table 2 The monitoring results of water of the drainage channel and the wetland. Indicator

Sample area Aa

Ph (dimensionless) COD (mg L−1 ) TP (mg L−1 ) TN (mg L−1 ) NO2 − (mg L−1 ) NO3 − (mg L−1 ) NH4 + (mg L−1 ) Ca2+ (mg L−1 ) Mg2+ (mg L−1 ) Na+ (mg L−1 ) K+ (mg L−1 ) Cl− (mg L−1 ) SO4 2− (mg L−1 ) HCO3 − , CO3 2− (mg L−1 ) TDS (mg L−1 ) Conductivity (ms/cm)

8.10 24.00 0.04 0.94 0.11 3.09 0.38 50.52 17.89 266.90 5.26 27.08 38.05 344.48 750.17 0.37

a b c d

± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

Sample area Bb

0.01 0.55 0.008 0.03 0.01 0.50 0.03 1.43 1.02 6.23 0.75 1.55 1.87 8.23 10.34 0.02

8.15 42.67 0.08 1.33 0.09 0.89 0.57 47.40 18.09 341.73 3.56 23.03 19.13 446.45 899.38 0.41

± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

0.01 1.18 0.003 0.08 0.004 0.05 0.02 1.56 1.13 7.32 0.54 1.67 0.98 8.87 11.03 0.03

Sample area Cc 8.03 21.33 0.05 0.76 0.07 0.68 0.42 48.08 17.35 262.79 3.97 18.98 20.04 391.95 763.17 0.36

± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

0.02 0.86 0.01 0.03 0.008 0.02 0.02 1.87 1.08 6.67 0.61 1.04 1.05 7.89 9.86 0.03

Sample area Dd 8.15 13.33 0.02 0.67 0.03 0.07 0.36 46.18 20.98 365.92 3.07 36.69 32.85 427.68 933.36 0.57

± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

The standard for Grade III water quality

0.01 0.87 0.01 0.02 0.007 0.005 0.03 1.66 1.43 7.02 0.46 1.77 1.73 8.05 10.37 0.04

6.5–8.5 20 0.05 1.0 0.15 20 – – – – – 250 250 – 450 2.5

Paddies drainage main channel. Wetland of South Lake. Center of South Lake. The North of Chagan Lake.

The water-soluble ash content was calculated with the following equation: Water-soluble ash content =

[(m1 − m0 ) − (m2 − m0 )] (m1 − m0 )

m0 – weight of the melting pot m1 – total weight of the melting pot and ash m2 – total weight of the melting pot and water-insoluble ash Twenty-five mL 100 g/L HCl was added into the residue which was left in the melting pot after the water-soluble ash content measurement, and the ash was washed into a 100 mL beaker, then heated in a water bath till the solution got clear. Heating continued for 5 min and the solution was filtered with an ashless filter paper till the solution got neutral (to 150 mL). The following steps were the same as for the water-soluble ash content measurement. Total weight of the melting pot and water-insoluble ash m2 was calculated. [(m1 −m0 )−(m2 −m0 )] Acid-soluble ash content = (m −m ) 1

0

m0 – the weight of the melting pot m1 – the total weight of the melting pot and ash m2 – the total weight of the melting pot and acid-insoluble ash (4) The determination methods of Cl− , SO4 2− , Na+ , K+ , Ca2+ , Mg2+ were the same as the methods for water and soil analyses. 3. Results and discussion 3.1. Choice of wetland desalination plants 3.1.1. The analysis of water and the bottom sediment The monitoring results of water and the bottom sediment are shown in Tables 2 and 3. As shown in Table 2, the main ions in the saline farmland drainage water were HCO3 − , CO3 2− , Na+ , Ca2+ , Mg2+ , K+ , Cl− and SO4 2− , and the pH in the lake was greater than 8 so that this lake water was alkaline. The detection of COD was above the value listed in the third category of the surface water quality standard (GB 3838 – 2002), which is 20 mg L−1 . The detections of TN and TP in area A, area C and area D were below the surface water quality standard for the third category, while the TN and TP contents of the sample

area B were above. The reason of this result is that there is greater density of wetland plants at the entrance of South Lake intercepting and capturing pollutants of drainage water coming from the area of the drainage channel, making the water with nitrogen and phosphorus pollutants stay in the South Lake for a long time and increasing the contents of TN and TP. The content of Mg2+ , Na+ and Cl− , TDS and the conductivity varied consistently. The indicators’ content increased first, then decreased, at last increased along the lake. The reason of this result is similar, there is greater density of wetland plants at the South Lake intercepting and capturing pollutants of drainage water coming from the area of the drainage channel, making the contents of some ions, TDS and conductivity higher. Pollutants, such as COD, P and N, from the drainage channel flowing through the South Lake wetland, South Lake to the North Lake, show the trend of increasing first and then decreasing. The main reason is the wetland at South Lake entrance. The water is discharged through the drainage channel to South Lake wetland, the pollutants are intercepted and captured by a large area of Typha spp., Phragmites communis, cattails and other wetland aquatic plants, and this forms a pollutant enrichment region around the wetland for a short period, leading to higher levels of pollutants. Then through the purifying effect of wetland and dilution of water by the widened water area, the concentrations begin to decline, and the monitoring indicators values keep quite low all the way to the Table 3 The monitoring results of the bottom sediment of the drainage channel and wetland. Indicator

Sample area Aa

pH (dimensionless) TP (g/kg) TN (g/kg) NH4 + (g/kg) Ca2+ (g/kg) Mg2+ (g/kg) Na+ (g/kg) K+ (g/kg) Cl− (g/kg) SO4 2− (g/kg) HCO3 − , CO3 2− (g/kg) TDS (g/kg) Conductivity (ms/cm)

8.27 1.25 1.52 0.72 0.31 0.08 0.22 0.04 0.26 0.32 0.05 1.89 0.27

a b d

± ± ± ± ± ± ± ± ± ± ± ± ±

Paddies drainage main channel. Wetland of South Lake. The North of Chagan Lake.

0.02 0.06 0.05 0.03 0.02 0.006 0.01 0.006 0.01 0.02 0.005 0.03 0.004

Sample area Bb 8.61 1.10 2.2 0.92 0.25 0.09 0.20 0.04 0.30 0.27 0.09 2.17 0.31

± ± ± ± ± ± ± ± ± ± ± ± ±

0.03 0.04 0.06 0.04 0.02 0.01 0.01 0.008 0.02 0.03 0.006 0.05 0.005

Sample area Dd 8.48 0.31 0.79 0.18 0.28 0.07 0.16 0.04 0.21 0.25 0.03 2.59 0.37

± ± ± ± ± ± ± ± ± ± ± ± ±

0.02 0.01 0.02 0.01 0.02 0.007 0.006 0.008 0.02 0.02 0.005 0.05 0.008

24

Y.N. Yang et al. / Agricultural Water Management 156 (2015) 19–29

Fig. 2. The monoclonal weight of the harvested part of different aquatic plants in the wetland: S-Ty, Typha spp. in South Lake; N-Ty, Typha spp. in North Lake; S-Ph, Phragmites communis in South Lake; N-Ph, Phragmites communis in North Lake; PhT, Phragmites japonica Steud. var. prostrata (Makino) L. Liu; Me, Medicago sativa Linn.

Fig. 3. The biomasses of the harvested part of the aquatic plants per unit area: S-Ty, Typha spp. in South Lake; N-Ty, Typha spp. in North Lake; S-Ph, Phragmites communis in Southlake; N-Ph, Phragmites communis in North Lake; PhT, Phragmites japonica Steud. var. prostrata (Makino) L. Liu.

North Lake. Gao et al. (2009) has found that submerged macrophytes in Donghu Lake could remove phosphorus, for example, total phosphorus (TP) removal rates of C. Demersum was 91.75% and 92.44% during the spring and autumn. Wang et al. (2014) studied that wetland plants had the capacity of removing nitrogen, for example, the total nitrogen (TN) removal rates of reeds and cattails were 90.4% and 82.3%. Carrie et al. (2011) researched that a large and diverse sample of wetland plant species can remove COD, and the COD removal rates of wetland plants were all above 60%. According to these research work and the distribution characteristics in Chagan Lake, the pollutants in the lake had been purified by the South Lake wetland aquatic plants to some extent along a longitudinal distribution, and the removal efficiency of COD, TP and TN respectively were 55.5%, 50.0% and 71.3%. The bottom sediment of Chagan Lake was weakly alkaline, and the variation trend of the contents of ions, conductivity and TDS was similar to that of the water because the wetland of South Lake intercepted part of the salt and deposited it at the bottom of the lake. Comparing Tables 2 and 3, it was found that the contents of TN, TP and salt in the bottom sediment were far higher than that in the water because of the deposition of pollutants at the bottom of the lake. Therefore, the ecological environment of Chagan Lake may be destroyed if a great quantity of drainage water flowed into it.

of Phragmites communis was three times the density of Typha spp. and both of their growth densities in South Lake were less than the growth of densities in North Lake. The growth density of Phragmites japonica Steud. var. prostrata (Makino) L. Liu was similar to the growth density of Typha spp. and Phragmites communis in South Lake, but the individual weight of Phragmites japonica Steud. var. prostrata (Makino) L. Liu was far less than Typha spp. and Phragmites communis. Thus, its unit area biomass, part of which could be harvested, was far less than that of Typha spp. and Phragmites communis. The individual weight of M. sativa Linn. was similar to that of Phragmites japonica Steud. var. prostrata (Makino) L. Liu. Moreover, Phragmites japonica Steud. var. prostrata (Makino) L. Liu and M. sativa Linn. are Carex taro swamp community. Due to the seasonal ponding at the ground surface, they also can be xeromorphic. They can only help desalinize the drainage water from large volumes. As shown in Fig. 3, the biomasses of Typha spp. and Phragmites communis were 14–15 kg/m2 and 6–7 kg/m2 , respectively, after July. The biomass of Potamogeton crispus reached its maximum of 4–5 kg/m2 in May. The biomass of L. minor L. was approximately 1 kg/m2 . Since L. minor L. is a floating and highly reproductive species, and the distribution of it can change very quickly, we usually measure its general biomass in July and August when there is more duckweed. The biomasses of Phragmites japonica Steud. var. prostrata (Makino) L. Liu and M. sativa Linn. were small. Therefore, the unit area biomasses, part of which could be harvested from Typha spp., Phragmites communis and Potamogeton crispus, were the biggest among these six aquatic plants. According to the results, Typha spp., Phragmites communis and Potamogeton crispus were considered to be the best desalination plants in Chagan Lake. Furthermore, Typha spp., Phragmites communis and Potamogeton crispus could grow well in the saline–alkali farmland drainage water because they could tolerate high salinity. Studies have showed that the growth limiting concentrations of Typha spp., Phragmites communis and Potamogeton crispus were 15 g/L, 30 g/L and 2.3 g/L (Wang et al., 2008), respectively, which was far higher than the concentration of 0.7–0.9 g/L in the farmland drainage water. Generally aquatic plants can only grow normally in water with the salinity of 0–0.5 g/L (Wang and Ji, 2007). Therefore, freshwater aquatic plants have a competitive advantage in non-saline environments, whereas the salt-tolerant plants like Typha spp., Phragmites communis can grow better in the saline environment. Since Typha spp. and Phragmites communis in Chagan Lake have grown over a long period of time, they have completely adapted to the higher concentration of saline environments.

3.1.2. Biomass analysis of wetland plants Plants can grow in a saline environment and absorb salt ions from the environment, so part of the salinity in the saline farmland drainage water can be removed by harvesting the plants. Therefore, the phytomass has a great influence on the desalination rate. A field experiment was conducted in the Qianguo irrigation district to choose the most efficient desalting plant by comparing the biomasses of Typha spp., Phragmites communis, Phragmites japonica Steud. var. prostrata (Makino) L. Liu, M. sativa Linn., L. minor L. and Potamogeton crispus. Fig. 2 shows that the individual weight of the wetland aquatic plants changed little after July and the main season of growth was before September. The reason of this result is that the growth of plants peak in August, and according to the plants growing cycle, individual weight of the plants increases first, then decreases (Deng et al., 2005). The comparative analysis showed that the individual weight of Typha spp. was much larger than the others and was approximately five times the weight of Phragmites communis in June and eight times the weight of Phragmites communis in July and September. The field measurements found that the growth density

Y.N. Yang et al. / Agricultural Water Management 156 (2015) 19–29

Fig. 4. The ash percentages of the different aquatic plants in the wetland: S-Ty, Typha spp. in South Lake; N-Ty, Typha spp. in North Lake, S-Ph, Phragmites communis in Southlake; N-Ph, Phragmites communis in North Lake; PhT, Phragmites japonica Steud. var. prostrata (Makino) L. Liu; Me, Medicago sativa Linn.; Le, Lemna minor L.; Po, Potamogeton crispus.

In this case, the salt concentration is not a major factor in restricting the aquatic plant growth. Typha spp. grew more easily where the water was abundant and the lake was deep, which caused the succession from the Phragmites communis community to the Typha spp. community in Chagan Lake. 3.1.3. The ash analysis of the wetland plants The plant biomass represents the plant quality per unit area, and the ash rate of the plants reflects how much mineral substance the plant absorbs and accumulates. Therefore, the product of biomass and ash rate can be chosen to represent the amount of salts contained in plants. A large product of the biomass and the ash rate means that a plant is better at removing salts. Comparative research was conducted on the ash rate of wetland aquatic plants including Typha spp., Phragmites communis, Phragmites japonica Steud. var. prostrata (Makino) L. Liu, M. sativa Linn., L. minor L. and Potamogeton crispus. The comparison results of the different wetland plants ash rates were shown in Fig. 4. The ash rates of plants were higher than 10%, for M. sativa Linn., L. minor L. and Potamogeton crispus. The ash rates of M. sativa Linn., L. minor L. were approximately 15.3%, which were the highest. The ash rate of Potamogeton crispus was approximately 10.5%. The ash rates of the others species were lower than 10%, including Typha spp., Phragmites communis and Phragmites japonica Steud. var. prostrata (Makino) L. Liu. The ash rate of Typha spp. was 6.5–7.5%, which was lower than that of Phragmites communis and Phragmites japonica Steud. var. prostrata (Makino) L. Liu. The ash rate of Phragmites communis and Phragmites japonica Steud. var. prostrata (Makino) L. Liu were 8.0–9.9% and 9.6%, respectively. The result showed that the ash content of the gramineous plants, such as Phragmites communis and Phragmites japonica Steud. var. prostrata (Makino) L. Liu, was higher than that of the typhacae plants. As a result, it was concluded that the ash rates of Typha spp. and Phragmites communis were higher than those in a non-saline fresh water environment, which were 5.4% and 5.8%, respectively. As shown in Fig. 4, the ash rates of Potamogeton crispus, M. sativa Linn. and L. minor L. were higher than those of Typha spp. and Phragmites communis. The products of the ash rate and the biomass of Potamogeton crispus, M. sativa Linn. and L. minor L. were still lower than those of Typha spp., Phragmites communis according to these plant experimental data. The biomass of L. minor L. was large in all seasons except for the frozen season, when it was difficult to find. The excess L. minor L. not only blocked the sunshine and made the

25

Fig. 5. The ash contents of the different aquatic plants per unit area: S-Ty, Typha spp. in South Lake; N-Ty, Typha spp. in North Lake; S-Ph, Phragmites communis in Southlake; N-Ph, Phragmites communis in North Lake; Po, Potamogeton crispus.

water become turbid, but it also caused a large number of organic chemical reactions and led to the death of fish and shrimp due to the lack of oxygen. It destroyed the ecological environment. Carex tato swamp communities such as Phragmites japonica Steud. var. prostrata (Makino) L. Liu and M. sativa Linn. can grow into sedge as a result of the seasonal surface water, which can also grow in a dry environment. They play a role in the desalination of the drainage water only when the yield of water is large. The ash rate of Potamogeton crispus was a little lower than that of Phragmites communis, Potamogeton crispus is not only a good greening material but is also the green feed for various types of livestock, poultry and herbivorous fish. It can increase the lactation yield and butter-fat content of dairy cattle and the laying rates, hatchability and fertility rates of poultry. Additionally, Potamogeton crispus also had some medicinal value, as it could clear heat and expel dampness (Wang and Zhang, 2005). Because the active period of Potamogeton crispus is from March to May, there would be low impact on the wetland ecosystems if they could be fished before June. Therefore, Typha spp., Phragmites communis and Potamogeton crispus were selected as the superior desalination plants based on the comparative results of the wetland plants’ biomasses and ash rates. 3.2. The desalination amount of the wetland aquatic plants The main ions in the saline farmland drainage water were HCO3 − , CO3 2− , Na+ , Ca2+ , Mg2+ , K+ , Cl− and SO4 2− (Burdick et al., 2001; Zhao and Fan, 2005; Omar, 2010; Xi et al., 2004). However, with the combined effect of the root systems of Typha spp. and Phragmites communis and the microorganisms in the aerobic, facultative anaerobic and anaerobic environments around the root systems, carbohydrate-containing HCO3 − and CO3 2− were likely to become carbon dioxide, which would be released into the air or be used by plants through photosynthesis. The removal of HCO3 − and CO3 2− did not relate to the wetland plants. Therefore, this paper focused on the desalination effects of Na+ , Ca2+ , Mg2+ , K+ , Cl− and SO4 2− . The plants in Chagan Lake can grow in the saline–alkali land as a result of their long-term adaptation. The Na+ , Ca2+ , Mg2+ , K+ , Cl− and SO4 2− have been detected in the ash of these plants, which shows that these plants in Chagan Lake have the ability of absorbing and removing salt. The total desalinization amount by harvesting wetland plants was calculated based on the ash determination results of Typha spp. and Phragmites communis in September, shown in Figs. 5 and 6. Fig. 6 shows that the Typha spp. harvested per square meter could remove approximately 234.8–262.6 g of salinity, in which

26

Y.N. Yang et al. / Agricultural Water Management 156 (2015) 19–29

Fig. 6. The soluble ash contents of the different aquatic plants per unit area: S-Ty, Typha spp. in South Lake; N-Ty, Typha spp. in North Lake; S-Ph, Phragmites communis in Southlake; N-Ph, Phragmites communis in North Lake; Po, Potamogeton crispus.

the water-soluble ash was 118.8–125.3 g/m2 and the acid-soluble ash was 218.5–243.9 g/m2 . The Phragmites communis harvested per square meter could remove approximately 215.0–352.4 g of salinity, in which the water-soluble ash was 54.2–73.5 g/m2 and the acid-soluble ash was 62.2–96.9 g/m2 . The Potamogeton crispus fished per square meter could remove 131.3 g of salinity, in which the water-soluble ash was 64.1 g/m2 and the acid-soluble ash was 119.4 g/m2 . 3.2.1. The analysis of the cation removal amount Fig. 7 shows that the cation contents removed by harvesting wetland plants in the saline farmland drainage water were calculated through the results of Ca2+ , Mg2+ , Na+ , and K+ in Typha spp. and Phragmites communis in September. The Typha spp. harvested per square meter could remove 21.1–31.1 g of Ca2+ , 8.7–9.3 g of Mg2+ , 18.8–19.9 g of Na+ , and 44.2–46.1 g of K+ , while the Phragmites communis harvested per square meter could remove 3.8–9.6 g of Ca2+ , 1.6–2.9 g of Mg2+ , 2.8–3.3 g of Na+ , and 5.0–15.3 g of K+ . The Potamogeton crispus harvested per square meter could remove 13.3 g of Ca2+ , 4.5 g of Mg2+ , 4.2 g of Na+ , and 15.7 g of K+ . The content of Na+ was the largest among the cations, based on the water quality monitoring results of the drainage water. Sodium salt is necessary for the growth of halophytes. Halophytes can reduce the cell water potential to regulate osmotic pressure by absorbing the sodium salt

Fig. 8. The anion contents of the desalination plants per unit area: S-Ty, Typha spp. in South Lake; N-Ty, Typha spp. in North Lake; S-Ph, Phragmites communis in Southlake; N-Ph, Phragmites communis in North Lake; Po, Potamogeton crispus.

in the environment. Halophytes also have a stimulation effect on non-halophytes that could increase their production (Brownell and Crossland, 1992). However, a large number of research studies have shown that sodium with a high concentration is considered to be the most harmful salinity to plants (Ashraf, 2009; Lin, 2007). Typha spp. was chosen to be the first plant for removing the salt from saline farmland drainage water because its desalination effect of Na+ was better than that of Phragmites communis and Potamogeton crispus. 3.2.2. The analysis of the anion removal amount As shown in Fig. 8, the anion contents removed by harvesting wetland plants in the saline farmland drainage water were calculated according to the results of Cl− and SO4 2− in Typha spp. and Phragmites communis in September. The Typha spp. harvested per square meter can remove 30.7–42.8 g of Cl− and 22.0–30.9 g of SO4 2− , while the Phragmites communis harvested per square meter can remove 13.5–17.1 g of Cl− and 20.0–25.7 g of SO4 2− . According to the results of Cl− and SO4 2− in Potamogeton crispus in May, the Potamogeton crispus fished per square meter can remove 5.3 g of Cl− and 12.3 g of SO4 2− . 3.2.3. The analysis of the desalination amount in the Chagan Lake wetland The annual drainage water of the drainage channel was 1.3 × 108 –2 × 108 m3 . According to the detection results, the TDS (salt content) average of the drainage water flowing into Chagan Lake was 750 mg L−1 (Fu et al., 2006), so the annual salt content of the drainage water was 0.98–1.5 × 105 t. According to field survey, the area ratio of reeds and cattails in North Lake of Chagan Lake is 4:1, while in South Lake of Chagan Lake, the area ratio is 13:7, the distribution area of reeds and cattails in Chagan Lake are shown in Table 4. Based on the unit area desalination amount of reeds and cattails and the removal of Ca2+ , Mg2+ , Na+ , K+ , Cl− , SO4 2− , it was calculated that the total desalination amount from harvesting the plants was 1.57–2.49 × 104 t per year in Chagan Lake (Table 5). The desalination amount by harvesting the aquatic plants was about 10–26% of the total salt of the Table 4 The distribution area of reed and cattail in Chagan Lake.

Fig. 7. The cation contents of the desalination plants per unit area: S-Ty, Typha spp. in South Lake; N-Ty, Typha spp. in North Lake; S-Ph, Phragmites communis in Southlake; N-Ph, Phragmites communis in North Lake; Po, Potamogeton crispus.

North Lake (km2 ) South Lake (km2 ) Total (km2 )

Reed

Cattail

44.34 11.38 55.72

11.09 6.13 17.22

Y.N. Yang et al. / Agricultural Water Management 156 (2015) 19–29

27

Fig. 9. Map of regional distribution of different wetland species in South Lake.

drainage water. The rates of removal of Na+ , K+ , SO4 2− , Ca2+ , Cl− and Mg2+ are shown in Table 6. The results showed that the K+ ion was important for osmosis regulation and resisting salt injury and could be accumulated in aquatic plants. It can be concluded that this method had an obvious desalination effect on the saline farmland drainage water. The desalination effects for K+ , SO4 2− , Ca2+ , Cl− and Mg2+ were better than for Na+ . The total rate of removal of the six ions Ca2+ , Mg2+ , K+ , Na+ , SO4 2− and Cl− was 9–15%. It was concluded that the aquatic plants in the wetland could remove part of the salt, but their removal capacity was limited because of the limited area of the wetland plants. If the removal efficiency of six ions reaches more than 80%, it is suggested that the existing wetland area be expanded by 6–9 times.

Table 5 The removal contents of different monitoring indicators. Indicator +

Na K+ SO4 2− Ca2+ Cl− Mg2+ Ash (W)a Ash (A)b a b

Water-soluble ash. Acid-soluble ash.

Content (×104 tons) 0.03–0.10 0.10–0.17 0.15–0.19 0.06–0.10 0.13–0.17 0.02–0.04 0.50–0.63 0.73–0.96

Chagan Lake wetland is national nature reserve, it plays an important role in tourism, fisheries and agricultural production in Northeast China. Due to the small amount of rainfall, higher level of evaporation, drought and some influence of human activities, Chagan Lake area reduces at a high speed. Only from 1984, when the project of the channel transferring water from Songhua River to Chagan Lake began, the wetland area of Chagan Lake recovered to the original state. Therefore, the saline water discharge using the wetland of Chagan Lake provides an adequate water supply for the Chagan Lake and plays a prominent part in the restoration and protection of Chagan Lake. South Lake and North Lake are basically covered with Typha spp., Phragmites communis and Potamogeton crispus. However, North Lake is so deep that aquatic plants have difficulty growing, especially in the middle of the lake, which is 4–6 m deep. Furthermore, the Typha spp. and Phragmites communis on the shore are weaker than those in South Lake because there are gravel mudstones in the sediment, and stormy waves are more prevalent. As a whole, North Lake is not suitable for the growth of aquatic plants, including Typha spp., Phragmites communis and

Table 6 The removal rates of the different ions. Ion

Na+

K+

SO4 2−

Ca2+

Cl−

Mg2+

Total salt

Rate (%)

3–4

100

50–67

17–28

54–74

18–21

10–26

28

Y.N. Yang et al. / Agricultural Water Management 156 (2015) 19–29

Potamogeton crispus. Therefore, additional constructed wetland is required if the removal rate of the six ions is needed to increase. The results showed that the desalination effect of Typha spp. was better than that of Phragmites communis and Potamogeton crispus. Typha spp. was recommended for removing the salt from the saline farmland drainage water. According to the calculations, 233–288 km2 of Typha spp. wetland should be constructed if the removal rate of the six ions is expected to be eighty percent. Taking the economic value of Potamogeton crispus into account, the area of wetland can be expanded appropriately, where Typha spp. is located around the lake and Potamogeton crispus is located in the middle of the lake. Fig. 9 shows the regional distribution of different wetland species. This study is based on the principle of saline–alkali soil improvement, using plants growing in the saline water habitats, and haloduric plants to desalt saline farmland drainage water. Previous research focused on the removal of the N, P elements, COD and some heavy metal ions, while our study focused on using wetland plants to desalt saline farmland drainage water. Yang et al. (2004) used a calculation method to estimate the mass fraction of soil salinity of reeds to be 2.0–7.0 g/kg, the average capacity of reeds absorbing salt 36.88 g/kg when the mass concentration of water salinity was 3.0–4.0 g/L. On the other hand, the mass fraction of soil salinity of cattails was 3.0–5.0 g/kg, the average capacity of cattails absorbing salt was 30.43 g/kg when the mass concentration of water salinity was 2.0–3.0 g/L. According to this experimental data, we could estimate the mass fraction of soil salinity of reeds and cattails were 1.9–2.6 g/kg, the average capacity of reeds absorbing salt was 56.24 g/kg when the mass concentration of water salinity was 0.7–1.0 g/L, while the average capacity of cattails absorbing salt was 16.80 g/kg when the mass concentration of water salinity was 0.7–1.0 g/L. It can be concluded that the desalination capability of reeds is higher than the cattails.

4. Conclusions The experimental results showed that Typha spp., Phragmites communis and Potamogeton crispus were the best plants for removing salt from saline farmland drainage in Chagan Lake by comparing the biomass, ash rate and ash composition of six aquatic plants, and 10–26% of the total salt of sodium was removed in the wetland by harvesting haloduric plants. If the removal efficiency of six ions reaches more than 80%, it is suggested to expand the existing wetland area by 6–9 times. This study also demonstrates that choosing Chagan Lake as intake area of saline farmland drainage water has beneficial and adverse effects. The beneficial effect is that saline farmland drainage water can be ecological water compensation for the Chagan Lake, while the adverse one is that the receiving water to the Chagan Lake is too much to be desalted by the plants. This may cause a lot of salt accumulating in the sediment of Chagan Lake and have an effect on the ecology of Chagan Lake, however, it is not clear what effect will it be on the ecology and how much will it affect the ecology. To answer these questions, a further monitoring study on the Chagan Lake must be conducted. For the agricultural water managers in northern China where the water resource is scarce, the research indicates that the saline farmland drainage is not only precious water resource, though it may cause secondary salinization as a pollution source. Therefore, it is very important to avoid the deterioration of the local ecological environment caused by high salinity in saline drainage water while improving the utilization of drainage water. The effective way to solve the above problems is to use aquatic and haloduric plants to desalinate and recycle the agricultural water to reduce drainage water wastage.

Acknowledgements This work was financially supported by Chinese Academy of Forestry’ Special Fund for Scientific Research in the Public Interest (201204201) and the Ministry of Water Resources’ Special Fund for Scientific Research in the Public Interest (200901038) and the Youth Special Foundation of IWHR’s Major Basic Research Program (1301). References Abulnour, A.G., Sorour, M.H., Talaat, H.A., 2002. Comparative economics for desalting of agricultural drainage water (ADW). Desalination 152, 353–357. Ashraf, M., 2009. Biotechnological approach of improving plant salt tolerance using antioxidants as markers. Biotechnol. Adv. 27 (1), 84–93. Bao, S.D., 2005. Agrochemical Analysis in Soil. China Agriculture Press, Beijing (in Chinese). Brownell, P.F., Crossland, C.J., 1992. The requirement for micronutrient by species having the C4 dicarboxylic photosynthesis pathway. Plant Physiol. 49, 794–799. Burdick, D.M., Buchsbaum, R., Holt, R., 2001. Variation in soil salinity associated with expansion of Phragmites australis in salt marshes. Environ. Exp. Bot. 46 (3), 247–261. Carrie, R.T., Paul, B.H., Otto, R.S., et al., 2011. Seasonal effects of 19 plant species on COD removal in subsurface treatment wetland microcosms. Ecol. Eng. 37 (5), 703–710. Darwisha, M.A., Al-Najem, N.M., Lior, N., 2009. Towards sustainable seawater desalting in the Gulf area. Desalination 235, 58–87. DB43/T 432-2009, Fresh Water Biological Resources Survey Technical Specification. Deng, F.T., Xu, S.J., Xu, X.H., et al., 2005. Selection and application of plants in constructed wetlands for pollution treatment of Dianchi Lake. Agric. Sci. Technol. 44, 299–303. EI-Zanati, E., EI-Khatib, M., 2007. Integrated membrane-based desalination system. Desalination 205, 15–25. Fan, Z.Q., 2003. Experimental Study of the Combination of Membrane Technology for Municipal Wastewater with Salt Retreatment. Harbin Institute of Technology, pp. 8–30 (in Chinese). Fu, C.P., Tang, Y.P., Xing, G.Z., et al., 2006. Nitrogen and phosphorus removal capacity of highly salty reuse water by wetland eco-restoring project. J. Irrig. Drain. 25 (4), 93–95. Gao, J., Xiong, Z., Zhang, J., et al., 2009. Phosphorus removal from water of eutrophic Lake Donghu by five submerged macrophytes. Desalination 242 (1–3), 193–204. GB 3838-2002, Environmental Quality Standard for Surface Water. Gilron, J., Chaikin, D., Daltrophe, N., 2000. Demonstration of CAPS pretreatment of surface water for RO. Desalination 127, 271–282. Lee, R.W., Glater, J., Cohen, Y., et al., 2003. Low-pressure RO membrane desalination of agricultural drainage water. Desalination 155, 109–120. Lin, J., 2007. A Study on Spatial Variability of Salinity and Nutrient in the Saline Soil and Element Chemical Characteristics. Xinjiang University, Urumqi, pp. 24–85 (in Chinese). Li, K., Wang, Y., Dou, S., 2009. Great project of land development and consolidation in the West of Jilin Province and engineering improvement of soda-saline soil. J. Jilin Agric. Sci. 34 (1), 20–24 (in Chinese). Li, S., 2013. Internal phosphorus of Chagan Lake Release and Impact on Eutrophication in Low Temperature Period. Jilin University, Jilin (in Chinese). Li, X.J., 2000. The alkili–saline land and agricultural sustainable development of the western Songnen plain in China. Sci. Geogr. Sin. 20 (1), 51–55. Lu, Y.Y., Hu, Y.D., Xu, D.M., et al., 2005. Optimum design for reverse osmosis seawater desalination system. Technol. Water Treat. 31 (3), 9–14. Lymbery, A.J., Doupe, R.J., Bennett, T., et al., 2006. Efficacy of a subsurface-flow wetland using the estuarine sedge Juncus kraussii to treat effluent from inland saline aquaculture. Aquac. Eng. 34 (1), 1–7. Malcolm, E., Sumner, R.N., 1998. Sodic Soils-Distribution, Properties, Management, and Environmental Consequences. Oxford University Press, New York. Omar, M.E., 2010. Efficiency of Duckweeds for Desalination and Treatment of Agricultural Drainage Water in Detention Ponds. Leibniz Centre for Agricultural Landscape Research, Müncheberg. Qin, L.J., Zhang, Y., Ren, L.J., et al., 2002. The soil salinization and the sustainable development of agriculture in the Qian Gorlos irrigation area, Jilin Province. Arid Zone Res. 19 (2), 52–55. Ren, W., Luo, T.B., Wang, B.J., et al., 2004. Biological improvement of saline and alkaline land in Xinjiang. Agric. Res. Arid Areas 22 (4), 212–215. Song, S.X., 2009. Yellow River Delta Re-saline Ecological Development Model of Analysis. China University of Petroleum, Beijing (in Chinese). Sun, K.J., 2014. Study on the balance between water supply and recharge in Chagan Lake. Jilin Water Res. (6), 1–5 (in Chinese). Wang, H.B., Zhang, S.H., 2005. The nutritional value, feed development and features research progress of Potamogeton crispus. Cereal Feed Ind. (7), 33–35. Wang, T.L., Su, F.L., Zhang, S., et al., 2008. Effects of salt stress on physiological properties of Phragmites australis and Typha latifolia. J. Shenyang Agric. Univ. 39 (4), 499–501. Wang, W., Ding, Y., Wang, Y.H., et al., 2014. The application and research progress of wetland plants in the nitrogen removal of constructed wetlands. Technol. Water Treat. 40 (3), 22–26.

Y.N. Yang et al. / Agricultural Water Management 156 (2015) 19–29 Wang, W.H., Ji, M., 2007. Comparison on salt tolerance of nine submerged macrophytes. J. Agro-Environ. Sci. 26 (4), 1259–1263. Xi, J.B., Zhang, F.S., Chen, Y., et al., 2004. A preliminary study on salt contents of soil in root-canopy area of halophytes. Chin. J. Appl. Ecol. 15 (1), 53–58. Xiong, W.C., 2005. Studies on Microwave Dielectric Behavior of Moist Salt Soil Inversion of the Moisture and Salt Content. Graduate University of Chinese Academy of Sciences, Beijing (in Chinese). Yang, F., Liang, Z.W., 2007. Strategic reflections on rice development in saline and alkaline area in West Jilin. North Rice (6), 7–12 (in Chinese). Yang, F., Zhang, G.X., Yin, X.R., et al., 2009. Spatial variability of soil salinity and alkalization and its correlation with microtopography in the west of Songnen plain. Sci. Geogr. Sin. 29 (6), 869–873.

29

Yang, F.Y., Wang, Z.C., Zhao, C.S., 2004. Effects of coordination of farming and fishery on environment of soil in saline–alkaline land. Ecol. Environ. 13 (1), 54–56. Yao, Y.Y., 2004. Integrated membrane process and related equipment for brackish water treatment. J. Sichuan Univ. (Eng. Sci. Ed.) 36 (2), 54–57. Zhang, T.Y., Liu, X.F., Qu, L., 2013. The characteristics and effects of the water quality of Lake Chagan affected by unicom water system. Jilin Water Res. (3), 1–8 (in Chinese). Zhang, Y.H., 2005. The desalination effect of halophytes in the saline soil. Gansu Agric. Sci. Technol. 3, 48–49. Zhao, K.F., Fan, H., 2005. Halophytes and Physiology of Adaptation to Saline Habitats. Science Press, Beijing, pp. 25–172. Zhao, K.F., Fan, H., Jiang, X.Y., et al., 2002. The role of halophytes in saline soil improvement. Chin. J. Appl. Environ. Biol. 8 (1), 31–35.