Introduction

The Chinese government has decided to popularize ecological circular agriculture throughout the country to recycle agricultural wastes, reduce mineral fertilizer input, and protect the environment. Agriculture in China has fully developed in recent years. The total grain yield in China exceeded 621 million tons in 2015 and has consecutively increased in the past 12 years (National Bureau of Statistics of the People’s Republic of China 2015). China is also a major producer of fruits and vegetables worldwide, with outputs reaching 260 and 700 million tons in 2014, respectively (Ministry of Agriculture of the People’s Republic of China 2014). Mineral fertilizers, especially nitrogen (N) fertilizer, have mainly contributed to this output. As reported, a threefold increase in N fertilizer application has contributed to about 70% increase in grain production in China since 1980, which has consumed huge amounts of oil resources but polluted the environment (Zhu and Chen 2002).

By contrast, anaerobic digestion has become increasingly popular worldwide to satisfy the growing concerns on energy supply and, more importantly, reduce organic wastes (Abraham et al. 2007; Arthurson 2009; Yu et al. 2010). China is a major developer of biogas engineering by anaerobic digestion. Approximately 600 million m3 of effluent could have been generated in rural areas in 2010 (Lu et al. 2012). However, the increasing popularity of anaerobic digestion has created another challenge that how to deal with large quantities of biogas slurry generated during this process. An 800 m3-volume biogas system needs to discharge 15 tons of biogas slurry daily. Therefore, this condition will become an environmental issue if improperly managed.

Studies showed that biogas slurry is a high quality nutrient material for crop growth and can be used as nitrogenous fertilizer because it contains large amounts of plant nutrients, such as P, K, and particularly N (Dahiya and Vasudevan 1986; Lu et al. 2012; Sheets et al. 2015; Tan et al. 2016). Biogas slurry probably also contains some micronutrients, auximones, B vitamins, and humic acid, which are beneficial to plant growth, can improve soil fertility, and can enhance farm production (Liu et al. 2009; Islam et al. 2010). The Chinese government has invested much to push biogas slurry use in crop cultivations for the reuse of agricultural waste and reduction of mineral fertilizer input. However, biogas slurry holds some chemical characteristics involving liquidity, high pH, high ammonium (NH4 +) content, and potentially high metal content (Hou et al. 2007; Lu et al. 2012; Svoboda et al. 2013), which differ with mineral fertilizers such as urea. Thus, studies must be performed to elucidate the effects of biogas slurry application on crop growth, yield, and environment and human health before its extensive use.

Researchers have studied biogas slurry utilization for crops, such as wheat (Garg et al. 2005), barley (Terhoeven-Urselmans et al. 2009), rice (Hou et al. 2007; Lu et al. 2012), leek (Elfstrand et al. 2007), maize (Svoboda et al. 2013), peanut (Zheng et al. 2016), and Chinese cabbage (Zhu et al. 2009). Most studies focused on the effects of biogas slurry on crop yield, soil properties, and fertility. Some researchers only selected environmental effects as their study point. However, the relatively integrated effects that involve crop growth, yield, nutrient use efficiency, agricultural product quality, and safety, as well as the potential environmental risks of applying biogas slurry to crop cultivation, are rarely investigated. Additionally, previous studies focused on cereal crops or conventional vegetables but not on aquatic vegetables, which account for a large proportion of the total vegetable planting area in southern China and require large amounts of nutrients during their entire growth.

Zizania aquatica Turcz. of Oryzeae has been cultivated for more than 2000 years and has historically been used mainly as an aquatic vegetable in China and some southeastern Asia such as Russia, Japan and Korea (Zhai et al. 2001; Guo et al. 2007). The smut fungus Ustilago esculenta causes enlargement of infected culms of Z. latifolia. The induced swollen culms are edible as a vegetable (Terrell and Batra 1982). There are two main types of cultivars cultivated in China at present. One is single-season Z. aquatica that can be harvested once each year, in the fall; the other is two-season Z. aquatica which can be harvested twice each year, in the fall and once the summer thereafter. The two-season cultivar is preferred by the Chinese farmers (Guo et al. 2007). By now, in China, Z. aquatica has been widely cultivated from northern Beijing to southern Guangdong province and Taiwan, and from eastern Sichuan province to Shanghai. The largest cultivated area in China is found in the region surrounding Tai Lake, Jiangsu and Zhejiang provinces (Chen 2002). It is estimated that more than 20,000 ha are under cultivation with Z. aquatica in Zhejiang province alone and that 500,000 tons of fresh product are harvested each year (Yu et al. 2003).

Improving the development of ecological circular agriculture in China will be more beneficial if biogas slurry, a byproduct of biogas engineering, can be used as mineral fertilizer (particularly N fertilizer) for Z. aquatica cultivation. Some farmers have replaced a partial amount of mineral fertilizer with biogas slurry during Z. aquatica growth in many places in China. However, a study should be conducted with the aim of assessing the plant growth responses and environmental impacts of applying biogas slurry in Z. aquatica cultivation. In addition, the feasibility of applying biogas slurry with irrigation should be studied. Field experiments were conducted in 2014 and 2015 in Jiaxing City, which is located in the Taihu Lake region of China. The study area is densely populated and has high potential for both cultivating Z. aquatica and developing biogas engineering because of large number of pig farms. In the present study, biogas slurry was applied to the plants at different rates and the yield, dry matter, and N use efficiency of Z. aquatica, N concentrations in floodwater and leachate, quality of agricultural product, and heavy metal contents in agricultural product and soil were measured. The results of this study would provide useful information that can be used to guide farmers in implementing best management practices in Z. aquatica cultivation.

Materials and methods

Experimental design

Field experiments were conducted in Jiaxing City, Zhejiang Province, China in 2014 and 2015. Jiaxing is located in the Hangjiahu Plain (30°50′N, 120°43′E) and is within the catchment of Taihu Lake. This area has a typical subtropical monsoon climate with an average annual temperature of 15–16 °C and an annual average rainfall of 1194 mm. Soil at the experimental site was identified as gleyed paddy soil [clay loam; mesic Mollic Endoaquepts (USDA/NRCS 1999)]. Prior to start of the field experiments, soil samples for site characterization were collected from the top 20 cm of the soil profile. The pH of the soil was 6.40 (in 1:2.5, soil to water) and organic C content, total N (TN), alkali-hydrolyzable N, Olsen P, NH4OAC-extractable K in the soil were 56.6, 3.53 g kg−1, 215, 5.89 and 119 mg kg−1, respectively.

Seven treatments based on the input N loading from N-fertilizer (NF) and/or biogas slurry (NB) were used in the experiments. The treatments were: no NF or NB (0N); N supplied completely by NF applied at traditional rate as practiced by farmers (1NF); from the N input equivalent to that received at 1NF half is supplied with NF and the other half with NB (0.5NFB); and N supplied only with NB at 1, 1.5, 2, and 2.7 times of N input equivalent to 1NF treatment (1NB, 1.5NB, 2NB and 2.7NB respectively). 1NF treatment received 450 kg N ha−1, 165 kg P2O5 ha−1 and 230 kg K2O ha−1 in first season (2014) and 485 kg N ha−1, 180 kg P2O5 ha−1 and 250 kg K2O ha−1 in second season (2015). Urea, single superphosphate and potassium chloride were used as the sources of N, P and K, respectively. From 0.5NFB to 2.7NB treatments the rates of P and K fertilizers were adjusted considering the P and K inputs from the biogas slurry. The fertilizers were applied in three splits per season (for N—35, 30, and 35%; for P and K—20, 30, and 50%) that correspond to three growth stages: seedling stage, early tillering stage, and early pregnant stage, respectively. Accordingly, fertilizer application was done on July 15, September 10, and October 9 in 2014 as for the first season, and on April 10, May 10, and June 4 in 2015 as for the second season. All mineral fertilizers were evenly dispersed by hand and biogas slurry was irrigated by 15 cm-diameter flexible pipes to each plot.

Biogas slurry was collected from the Shuangqiao Black Pig Farm. Wastewater from the solid–liquid separation of pig farm waste (i.e., a mixture of pigs’ dejecta and the wastewater from piggery cleaning) was anaerobically digested for 20–30 days or longer time under normal temperature digestion condition to produce biogas in a methane fermentation tank. The biogas slurry was then discarded and stored in a pond for experimental usage. Before irrigation, we collected two parallel samples of biogas slurry in plastic bottles after stirring biogas slurry body as evenly as possible. The chemical properties of biogas slurry for each irrigation treatment are shown in Table 1. The irrigation capacity of biogas slurry was defined as N rate supplied for each treatment divided by TN concentration in biogas slurry. The biogas slurry was transported by a collection vehicle with a storage tank to the test site for fertilization. A 15 cm-diameter flexible pipe was used to connect the storage tank and each plot. The treatments were applied in triplicates. Twenty-one plots each with an area of 84 m2 (12 m × 7 m) were arranged in a randomized complete block design. The field levees of each plot were 35–40 cm high above the soil plane in the plot. The plastic film was buried in the ground at a vertical depth of 30 cm to avoid lateral seepage between two adjacent plots.

Table 1 Chemical property of biogas slurry irrigated (mg L−1)
Full size table

Two-season Z. aquatica variety Zhe911 was selected as the experimental material. It is widely cultivated in the Zhejiang and Jiangsu Provinces of China. The plants were transplanted on July 1, 2014. The distance between rows and hills was 70 cm × 80 cm. Floodwater level was maintained at a depth of 15–20 cm during the growth stage in each plot. However, floodwater level was reduced to 5–10 cm at the maximum tillering stage, which helped to inhibit excessive tillering that can’t develop the field to a great degree. The swollen culms were harvested by hand from October 20 until November 10, 2014. Subsequently, the above-ground parts were removed when the weather was sufficiently cool to prevent crop regrowth. It was regrown in late March 2015 when the weather was sufficiently warm. It was harvested from June 16 to July 15, 2015.

Sampling and measurements

Eight representative hills of Z. aquatica plants in inner rows of each plot were selected to determine yield in 2014 and 2015 growing season. The fresh swollen culms were harvested and weighed for yield definitions. The above-ground parts (including leaf, stem and swollen culm) of four hills of plants chosen for yield in each plot were collected at harvest time. These above-ground parts were then oven-dried at 75 °C until a constant weight was reached for dry matter measurements. Subsequently, the samples were ground for N concentration measurements. The samples of fresh swollen culms removing shell (as edible part) were collected, some of which were oven-dried at 75 °C until a constant weight was reached, and some of which were used for quality analysis of agricultural products. The dried samples of non-shell swollen culms were ground for N concentration measurements. In 2015 season, some milled samples were sieved through a 0.149 mm mesh to use in the analyses of potentially toxic trace elements to confirm whether the edible part of Z. aquatica was polluted by heavy metal in biogas slurry after a relatively long irrigation period (in 2014 and 2015 growing seasons).

A composite soil sample mixed by five subsamples at depth of 0–20 cm was collected in each plot after harvest of Z. aquatica in 2015 with the same purpose as that mentioned above. Naturally air-dried and homogenized samples were ground with an agate mortar, passed through a 150 μm nylon sieve (100 mesh), and stored in closed polyethylene bags for analyses of potentially toxic trace elements.

The floodwater samples in plots were collected 1 day after each fertilization. Collection was repeated at 3 day intervals until similar NH4 +–N contents were obtained among treatments. The water samples were filtered with general qualitative filter paper and subsequently frozen in −20 °C for TN, ammonium–N (NH4 +–N), and nitrate–N (NO3 −–N) concentrations analyses.

PVC tubes with porous at one end were installed in each plot at a depth of 75 cm to collect percolating water since the groundwater depth is ~1 m in this region. The percolating water was pumped out several times with a hand pump before sample collection. Leachate samples were collected at 7 or 10 day intervals after fertilizer application during the growing season. The water samples were filtered with general qualitative filter paper and immediately stored in a freezer at −20 °C until the TN, NH4 +–N, and NO3 −–N concentrations analyses.

Chemical analysis of samples

Milled plant samples were digested with H2SO4–H2O2. Subsequently, the N concentration was determined by the Kjeldahl method. Vitamin C (Vc) in non-shell swollen culms was extracted with 2% oxalic acid and analyzed by the 2,6-dichlorophenolindophenol sodium salt titrimetric method (Liu et al. 2008). The total soluble sugar content was determined with a colorimetric method (Laurentin and Edwards 2003). Anthrone (0.15%) in ethanol was added to suitable aliquots of the extract or the standard sugar solution, mixed, and incubated for 15 min in a water bath at 90 °C. The absorbance was determined at 620 nm with a spectrophotometer (DR 5000 Spectrophotometer by HACH). The protein was analyzed based on a method in GB 5009.5-2010 (National Food Safety Protein in Foods). The milled samples of non-shell swollen culms were digested with CuSO4–K2SO4–H2SO4. The N concentration was subsequently determined by the Kjeldahl method. The protein content was obtained by multiplying the N concentration by 6.25, which is a coefficient used to convert N content to protein. The nitrate (NO3 −) content in non-shell swollen culms was determined according to the method as described by Miranda et al. (2001). The homogenate of the samples in hyperpure water was mixed with 5% salicylic acid in the sulfuric acid solution and incubated with 1.8% sodium hydroxide for 20 min at room temperature. The absorbance was read at 410 nm. The level of nitrate in all samples was calculated according to standard curves.

The concentrations of TN, NH4 +–N, and NO3 −–N in floodwater, percolating water, and biogas slurry were measured as the followings: The TN concentration was determined with a spectrophotometer at 220 and 275 nm (DR 5000 Spectrophotometer, HACH) after alkaline persulfate digestion; The NH4 +–N content was measured by an indophenol blue colorimetric method; The NO3 −–N concentration was analyzed with a spectrophotometer at 220 and 275 nm (DR 5000 Spectrophotometer, HACH). The total P content in biogas slurry was analyzed by the molybdenum blue method after persulfate digestion. The total K concentration in biogas slurry was determined by flame atomic absorption spectrophotometry (FP640; Shanghai Spectrum Instruments Co., Ltd, China). The pH of biogas slurry was determined by using a pH meter (FE 20; Mettler Toledo).

The potentially toxic trace elements (Cu, Zn, Pb, Cr, Cd, As, and Hg) in biogas slurry, soil, and non-shell swollen culms were measured by inductively coupled plasma mass spectrometry (Agilent Technologies, Inc., USA) after HNO3 digestion (Lu et al. 2012; Duan et al. 2012).

Statistical analysis

Data were statistically analyzed with SPSS version 17.0. One-way ANOVA with the LSD post hoc test was performed to determine differences among different treatments. Significant differences at P < 0.05 are indicated by different letters. Figures were plotted with Sigma Plot version 10.0.

Results

Yield and dry matter accumulation of two-season Z. aquatica

The Z. aquatica yield was significantly reduced when no N fertilizer was supplied in both growing seasons. No significant differences were observed between 1NF versus the 0.5NFB, 1NB, 1.5NB, and 2NB treatments. However, the yield was significantly reduced by almost 10% in 2.7N compared with that in 1NF in 2014 (Fig. 1a). Further, the yield was reduced by 48.8% in 2.7NB and by 26.2% in 2NB compared with that in 1NF in 2015 (Fig. 1b).

Fig. 1
figure 1

Yield of Z. aquatica in the 2014 (a) and 2015 (b) growing seasons. Error bars represent the standard deviation; different lowercase letters indicate significant differences (P < 0.05) among treatments

Full size image

The lack of N input significantly decreased the dry matter accumulation in above-ground parts of Z. aquatica by 24.0–40.3% in 2014 and 16.4–48.8% in 2015. No obvious differences in dry matter development were observed between 1NF versus 0.5NFB, 1NB, and 1.5NB in both growing seasons. However, the dry matter was significantly reduced by 9.44 and 21.4% in 2NB and 2.7NB in 2014 and by 21.1 and 38.6% in 2015, respectively (Fig. 2a, b).

Fig. 2
figure 2

Dry matter accumulation in above-ground parts of Z. aquatica plants in the 2014 (a) and 2015 (b) growing seasons. Error bars represent standard deviation; different lowercase letters indicate significant differences (P < 0.05) among treatments

Full size image

N accumulation and physiological N use efficiency (PNuE) of two-season Z. aquatica

The N accumulation in above-ground parts was significantly restrained by 16.6% for 2.7NB, whereas 0N was 103% lower than 1NF in 2014. However, a significant decrease was found in the 2NB and 2.7NB treatments in 2015 by 13.8 and 30.5%, respectively, except for 0N with a decrease of 59.2% compared with 1NF (Fig. 3a, b).

Fig. 3
figure 3

N accumulation in above-ground parts of Z. aquatica plants in the 2014 (a) and 2015 (b) growing seasons. Error bars represent standard deviation; different lowercase letters indicate significant differences (P < 0.05) among treatments

Full size image

The PNuE of plants is defined as the biomass accumulation relative to the N accumulation (Shi et al. 2010). The absence of N application markedly increased the PNuEs of above-ground parts of Z. aquatica by 16.3–22.4% in 2014 and by 20.3–29.7% in 2015 compared with other treatments. PNuEs were not obviously affected in the 0.5NFB, 1NB, and 1.5NB treatments as compared with the 1NF treatment. However, the 2NB and 2.7NB treatments decreased the PNuE by 4.65 and 5.91% in 2014 and by 8.56 and 11.8% in 2015, respectively (Fig. 4a, b).

Fig. 4
figure 4

Physiological N use efficiency in above-ground parts of Z. aquatica plants in the 2014 (a) and 2015 (b) growing seasons. Physiological N use efficiency in the above-ground parts of Z. aquatica plants: dry matter accumulation (kg ha−1) divided by the N accumulation (kg ha−1) in above-ground parts of Z. aquatica plants. Error bars represent standard deviation; different lowercase letters indicate significant differences (P < 0.05) among treatments

Full size image

N concentration in floodwater

Immediately after the application of mineral N fertilizer or biogas slurry in both growing seasons, the TN and NH4 +–N concentrations in floodwater (Fig. 5a–d) were higher than 30 and 25 mg L−1, respectively, which are the critical levels set in the National Integrated Wastewater Discharge Standard of China (GB 8978-1996). However, the treatments involving biogas slurry application had obviously higher concentrations than the 1NF treatment in most cases, especially for 1NB, 1.5NB, 2NB and 2.7NB treatments. Moreover, TN and NH4 +–N concentrations showed increased tendency with increasing of biogas slurry input. The TN (or NH4 +–N) content decreased below critical level on the 7th, or 10th, or 13th day, which indicated that the high N concentrations in floodwater did not immediately reach critical levels. Although the application of more biogas slurry increased the NO3 −–N content in floodwater, the maximum was only 10.6 mg L−1 in 2014 or 13.2 mg L−1 in 2015. No critical level of NO3 −–N concentration is set in the National Integrated Wastewater Discharge Standard of China (GB 8978-1996). But, its maximum in floodwater was far below the national standards for discharge concentrations of TN and NH4 +–N (Fig. 5e, f).

Fig. 5
figure 5figure 5

TN, NH4 +–N, and NO3 −–N concentrations in floodwater after biogas slurry application in the 2014 (a, c, e) and 2015 (b, d, f) growing seasons. Columns from left to right in each table above the variation curve correspond to the sampling date order. Different lowercase letters in the same column indicate significant differences (P < 0.05) among treatments. Abbreviation ns at the top row of the table indicates no significant differences (P < 0.05) among treatments

Full size image

N concentration in percolation water

There were no obvious differences in N concentrations (TN, NH4 +–N, and NO3 −–N) in percolating water in most instances between treatments received biogas slurry and 1NF, which implied biogas slurry application showed the similar effects on the N contents in percolating water in comparison with mineral N fertilizer application. Interestingly, N applications by mineral N fertilizer or biogas slurry did not significantly increase the N content in the percolating water of this area (Fig. 6a–f).

Fig. 6
figure 6figure 6

TN, NH4 +–N, and NO3 −–N concentrations in percolation water during different growth stages in the 2014 (a, c, e) and 2015 (b, d, f) growing seasons. Different lowercase letters in the same column above sampling date indicate significant differences (P < 0.05) among treatments. Abbreviation ns at the top row of table indicate no significant differences (P < 0.05) among treatments. Arrow with day on the top indicated the day of fertilization event after planting

Full size image

As shown in Fig. 6a–f, the TN, NH4 +–N, and NO3 −–N concentrations were in the ranges of 1.16–4.59, 0.2–2.10, and 0.51–2.03 mg L−1, respectively, in the 2014 and 2015 growing seasons for all treatments. However, the NH4 +–N contents exceeded 0.5 mg L−1 (a critical level according to the National Groundwater Quality Standard of China or GB/T 14848-1993) during most of the Z. aquatica growth stages. By contrast, the NO3 −–N concentrations were much lower than 5.0 mg L−1 (a critical level according to the National Groundwater Quality Standard of China or GB/T 14848-1993). This critical level was not observed for the TN concentration in percolating water.

Agricultural commodity quality and heavy metal risk in non-shell swollen culms and soil

As shown in Table 2, the Vc content in non-shell swollen culms of Z. aquatica increased by 12.6–17.8% in 2014 and 13.6–17.0% in 2015 after biogas slurry treatment compared with 1NF. However, increasing the biogas slurry input did not significantly increase Vc content. Obviously, the increased protein content was observed in 2NB and 2.7NB, particularly 2.7NB, which was probably caused by increasing N application rate. No significant differences in the total sugar content were found among treatments. Moreover, the biogas slurry application did not increase the nitrate content in non-shell swollen culms of Z. aquatica.

Table 2 Quality of non-shell swollen culm in 2014 and 2015 growing season
Full size table

There were no significant differences in Cu, Zn, Pb, Cr, Cd, As, and Hg concentrations in non-shell swollen culm or soil between 0N and the treatments supplied with N fertilizer. Moreover, biogas slurry application did not increase the heavy metal contents, regardless of the application rates. Heavy metal contents in non-shell swollen culm were below the critical values stipulated in the National Maximum Levels of Contaminates in Foods of China (GB2762-2005), the National Maximum Levels of Cu in Foods of China (GB13106-91), and the National Maximum Levels of Zn in Foods of China (GB15199-94). The detected heavy metal contents in soil were below the critical values stipulated in the National Soil Environmental Quality Standard of China (GB15618-1995) (Tables 3, 4).

Table 3 Heavy metal content in non-shell swollen culm in 2015 growing season (mg kg−1)
Full size table
Table 4 Heavy metal content in soil in 2015 growing season (mg kg−1)
Full size table

Discussion

Excessive biogas slurry application suppressed growth, yield formation, and PNuE

As a kind of N fertilizer, biogas slurry could completely or partially replace mineral N fertilizer for the growth of Z. aquatic. Similar findings were reported in other crops, such as rice, wheat, maize, and komatsuna (Hou et al. 2007; Lu et al. 2012; Svoboda et al. 2013). However, excessive use of biogas slurry adversely affects the Z. aquatica yield, dry matter, and N accumulation in above-ground parts. In addition, the growth suppression was more serious when continuously applied in the second growing season (Figs. 1a, b, 2a, b, 3a, b). Moreover, the plant height, leaf length, and stem length also decreased because of excessive application (unpublished). Similar results have been observed in other crops, such as bhendi and fodder maize, although these crops showed various responses to different rates of biogas slurry application (Geeta and Sreenivasa 2002; Islam et al. 2010).

NH4 + is an inorganic nitrogen source that can be deleterious to the growth of plant species when absorbed as the sole N source (Schortemeyer et al. 1997; Kotsiras et al. 2005; Dai et al. 2008). Rice is an NH4 +-tolerant species that can also be negatively affected by high NH4 + concentrations in solution (Balkos et al. 2010). Chen et al. (2013b) found that different rice varieties had diverse tolerance levels to high NH4 + concentrations in solution. The levels of NH4 + in soils frequently reach the critical amount, which negatively affects plant growth. These negative effects are manifested as stunted root growth, yield suppression, and leaf chlorosis (Britto and Kronzucker 2002). We collected biogas slurry samples from eight pig farms in Jiaxing City and found NH4 +–N is a dominant form of N that accounts for 77–93% of the TN in biogas slurry (data unpublished). These levels were similar with those obtained in other reports (65–90%) (Chen et al. 2013a; Svoboda et al. 2013). In the present study, as much as 300 mg L−1 NH4 +–N was present in floodwater after biogas slurry application. Moreover, NH4 +–N above 150 mg L−1 existed in floodwater for 6 days or more (Fig. 5c, d). This phenomenon may lead to high NH4 + concentrations around the Z. aquatica root. Additionally, Li et al. (2012a) confirmed that the growth and lateral root development of Arabidopsis was considerably influenced by NH4 + toxicity when leaves were exposed to high NH4 + levels as compared with the root. In the present study, partial stems of Z. aquatica were exposed to high NH4 + levels for several days, which might have contributed to growth depression. Chen et al. (2013b) suggested that the decreased N use efficiency in plants may be closely associated with the increased futile NH4 + cycling in roots and NH4 + toxicity, which might be the principal reason for decreased PNuE levels in above-ground parts of Z. aquatica.

Biogas slurry application has been widely recognized to improve the quality of crops or fruits (Liu et al. 2009; Xue et al. 2012). The preset study demonstrated that biogas slurry application significantly increased the Vc content in non-shell swollen culms of Z. aquatica but did not adversely affect the nitrate content (Table 2). In addition to N, P, and K, biogas slurry contains other important nutritional substances, such as carbohydrates, amino acids, microelements, hormones, and crude proteins, which might play an important role in improving product quality (Liu et al. 2009; Islam et al. 2010).

Environmental effects of biogas slurry applied on Z. aquatica

The environmental effects after application of biogas slurry have gained an increasing amount of attention. In this study, biogas slurry irrigation and mineral N fertilizer application showed the similar effects on the N contents in percolating water (Fig. 6a–f); these results are similar to the reports of Lu et al. (2012) and Chen et al. (2013a). The soil characteristics in test area of purple clay-based paddy soil may be closely related to our results because high soil density and poor water penetration capability were observed (Li et al. 1984).

High concentrations of NH4 +–N in floodwater could be a potential risk to ambient water bodies when runoff events occur (Chen et al. 2016). In the present study, high TN and NH4 +–N concentrations of up to ~300 mg L−1 were observed in floodwater after biogas slurry was irrigated; these concentrations increased with more input and reduced to the safe level after 10–13 days or longer time (Fig. 5a–d). Runoff events occurring during this period can introduce considerable risks to the water body. However, a high ridge of field is usually built to keep high water level in field for aquatic vegetable growth. For example, farmers usually build 35–40 cm or higher ridges for planting Z. aquatica in fields. Therefore, the runoff event does not normally occur only if 150–200 mm of precipitation occurrs within a short period of time. In our study, runoff events did not occur in the 2014 and 2015 growing seasons.

The rapidly decreasing NH4 +–N content in floodwater could be mainly attributed to ammonium (NH3) volatilization and nitrification–denitrification processes. Previous studies showed that the application of biogas slurry produced 2–4 times more NH3 and nitrous oxide (N2O) than the mineral fertilizers, which was probably aggravated by applying more biogas slurry (Hou et al. 2007; Jacobi et al. 2009; Chen et al. 2013a). Furthermore, N uptake by crops also can decrease the NH4 +–N content in floodwater. The study suggests that NH3 volatilization in paddy fields was considerably reduced when the mineral N fertilizer was applied at the elongation stage as compared with the early tillering stage; this trend is attributed to the strong ability of N uptake by rice plants at elongation stage (Cassman et al. 1998). In our study, the required time to reduce NH4 +–N to safe levels was relatively long, with more biogas slurry input. The main reasons for this trend were (1) the large quantity of irrigation and the increased NH4 +–N concentration in floodwater and (2) the decreased N uptake of Z. aquatica plants by excessive use of biogas slurry.

Volatilized NH3 can return to land and water by the dry deposition of NH3 or the dry–wet deposition of NH4 +; these phenomena can contribute to acidification and eutrophication of water bodies, thereby causing undesirable changes in natural ecosystems (Bouwman et al. 2002). N2O is a major greenhouse gas that is 250 times more effective as an absorber of infrared radiation than CO2 for a 100-year time frame (Ruser and Schulz 2015). Therefore, simple and effective technologies should be developed to limit the NH3 and N2O emissions induced by irrigating biogas slurry.

The heavy metal pollution in agricultural commodities and soil has become a growing concern. In this study, the heavy metal content in the non-shell swollen culms and soil did not increase after applying the biogas slurry (Tables 3, 4). Similar findings were reported by Lu et al. (2012) and Montemurro et al. (2008). The first possible reason for this trend is that the heavy metal concentrations are low in biogas slurry, which was also reported by Arthurson (2009) and Lu et al. (2012). Furthermore, Marcato et al. (2009) reported the lower bioavailability of Cu and Zn in anaerobically digested pig slurry than in the raw slurry. Second, the heavy metal concentrations of biogas slurry were significantly lower than the background levels in the soil. Finally, high concentrations of organic matter in biogas slurry can form complexes with heavy metals, thereby decreasing the available amount for plant uptake (Lo et al. 1992; Marcato et al. 2009). However, the cumulative long-term effect of heavy metals in agricultural commodities and soil must be further monitored to ensure environment and human safety.

Confirming NH4 + content in biogas slurry for its improved reasonable use in crop cultivation

The Chinese government has encouraged the recycling of biogas slurry within agriculture system to develop ecological circular agriculture and reduce mineral fertilizer application rates (Li et al. 2012b; Qin 2015). However, the problem is that it is difficult for farmers to determine the N content in biogas slurry applying for crops or other plants growth. High NH4 + content in biogas slurry and its excessive application can damage the plants. Although, previous studies have recommended a total range of application rates or one certain rate on selected crops (Wu et al. 2011; Lu et al. 2012; Zheng et al. 2016). The application rate of biogas slurry is still difficult to confirm because of large variations in the N concentration. In the present study, 395–1521 mg L−1 TN and 301–1141 mg L−1 NH4 +–N were detected in biogas slurry (Table 1). Additionally, biogas slurry samples were collected from eight pig farms in the Jiaxing area; the observed TN content ranged from 260 to 1358 mg L−1, with 236–1196 mg L−1 of NH4 +–N accounting for 77–93% of the TN (data unpublished). Lu et al. (2012) reported TN and NH4 +–N concentrations of 531–1117 and 492–733 mg L−1, respectively. NH4 +–N is the primary N form in biogas slurry because methane is extracted, whereas N in the substrate is conserved in biogas residues, thereby increasing the proportion of N in NH4 +–N and reducing the C/N ratio (Gutser et al. 2005). The strong redox potential and alkaline conditions in biogas slurry are also important contributors (Lu et al. 2012). Therefore, the optimal N concentrations should first be understood before biogas slurry can be reasonably and safely used for crop cultivation.

TN or NH4 +–N concentrations in water are generally measured by spectrophotometry in the laboratory. The determination of TN content is complicated because of the high-temperature and high-pressure digestion processes and the spectrophotometry at UV–light. However, NH4 +–N is measured via an indophenol blue colorimetric method by spectrophotometry at the visible light range and does not need to undergo digestion (Peng et al. 2011). Moreover, NH4 +–N is the most prevalent form of N in biogas slurry, and a large amount of N levels is present. Therefore, the development of simple and quick methods to determine the NH4 +–N content in biogas slurry should be feasible and beneficial to quantify biogas slurry irrigation rates. At present, semi-quantitative methods are used for NH4 +–N determination in related commercial products. Nevertheless, the low-concentration of NH4 +–N (0–100 mg L−1) in water was analyzed but was found inconvenient for high NH4 +–N measurement in biogas slurry. To promote the reasonable use of biogas slurry for crop cultivation, semi-quantitative or precise methods should be developed to determine the NH4 +–N concentration in biogas slurry.

Conclusion

Biogas slurry was used as N fertilizer to partially or completely substitute mineral N fertilizer and satisfy two-season Z. aquatica (an aquatic vegetable) growth, including the yield formation, dry matter development, N uptake, and N utilization. Nevertheless, excessive application (2 or 2.7 times higher N in the biogas slurry relative to the mineral N fertilizer in our study) suppressed normal growth, which was probably caused by NH4 + toxicity. The application of biogas slurry improved the quality of non-shell swollen culm but maintained safe concentrations of nitrate and heavy metal content. Furthermore, biogas slurry application did not cause heavy metal pollution in the soil or N pollution in leakage water. However, the N concentrations in floodwater are a major threat to ambient water bodies after the application of biogas slurry as compared with mineral N fertilizer. High NH4 +–N and the large variation of N content in biogas slurry impedes its reasonable use as nitrogenous fertilizer for crop cultivation. Thus, semi-quantitative or more precise methods are necessary, especially to determine the NH4 +–N concentration in biogas slurry, given its large proportion of TN content and relatively simple measurement.