Introduction

Super-intensive shrimp farming has been a great success (Holl et al. 2011; Ge et al. 2016a; Pungrasmi et al. 2016). However, the environmental impacts of aquaculture activities in recent years have drawn great attention (Xu and Pan 2012; van Rijn 2013; Zhu et al. 2015). Ammonia and nitrite in aquaculture wastewater cannot only pollute the environment, but also can cause serious damage to shrimp. Reducing the negative influences of aquaculture on environment is key to ensuring the long-term sustainability and the further development of the industry (Vinatea et al. 2010; Furtado et al. 2015). Therefore, nitrosobacteria and nitrifying bacteria filtration in recirculating aquaculture system (RAS) have been employed to control ammonia (Guerdat et al. 2010; Kuhn et al. 2010a). However, nitrate nitrogen accumulation can also reduce survival and growth of shrimp (Kuhn et al. 2010b; Furtado et al. 2015). Moreover, nitrate in the wastewater without treatment may lead to environmental pollution. However, the removal of nitrate in wastewater is complicated and costly (Cahill et al. 2010; Furtado et al. 2015).

Seaweeds are another promising method to control inorganic nitrogen (Baloo et al. 2014; Ge et al. 2016a). Ulva prolifera is a cosmopolitan green seaweed (Sun et al. 2015) which can use multiple nitrogen sources (Sun et al. 2015; Li et al. 2018a) and acclimatize to the changes of temperature by morphology-driven physiological and biochemical variation (Gao et al. 2016). These characteristics allow U. prolifera to adapt to the environment of shrimp farming (such as high water temperature, nutrient enrichment, and so on) and act as a biofilter to control ammonia concentration (Baloo et al. 2014). Therefore, green seaweeds such as U. prolifera have drawn great attention (Liu et al. 2009; Yabe et al. 2009). The dry matter of U. prolifera has high carbohydrate (62%) and protein (27%) and low lipid (0.3%) contents (Li et al. 2015; Shao et al. 2017). The polysaccharide from Ulva spp. has biological activities such as immunomodulatory (Kim et al. 2011; Akbary and Aminikhoei 2018) antioxidant (Shi et al. 2017; Li et al. 2018b), antimicrobial (Berri et al. 2016; Shao et al. 2017), and anticancer (Jiao et al. 2009; Murphy et al. 2014) activities. In addition, Ulva co-cultured with shrimp can convert ammonia, nitrite, and nitrate into biomass, and the biomass can be harvested and processed into animal feed or for human consumption (Wang et al. 2007a, b; Zhang et al. 2011; Bikker et al. 2016; Qiu et al. 2018). However, inorganic nitrogen concentration in wastewater is high and seaweed can only partially remove the inorganic nitrogen. In the present study, U. prolifera was selected as a biofilter. The appropriate quantity of integrating U. prolifera with white shrimp was determined first, and then, the effects of integrating U. prolifera with white shrimp (500 shrimp m−3) at varying water exchange levels on water quality and shrimp growth performance were explored. The growth performance of U. prolifera was also monitored.

Materials and methods

Source of seaweed and shrimp

The green seaweed U. prolifera was provided by Xiangshan Xuwen Seaweed Development Co., Ltd. in Zhejiang province. The seaweed was cultured in sterile culture and the branch length was 2.51–3.03 cm. Specific pathogen-free L. vannamei (PLs 5) was obtained from a commercial breeding station (Chia-Tai, Hainan) and the PLs were cultured at a stocking density of 1000 shrimp m−2 in a nursery tank until the shrimp reached a mean weight of 3.45 g.

Experimental set up

Preliminary experiment: Effect of U. prolifera on water quality of white shrimp reared in the zero water exchange system zero exchange water

To determine appropriate range of U. prolifera in the shrimp culture system, a preliminary trial of controlling water quality by U. prolifera was conducted. Seventy-five shrimp were stocked at a density of 500 shrimp m−3 in PVC tanks (r = 0.245 m, h = 1 m, water depth = 0.8 m). The initial weight of shrimp was 3.01 ± 0.11 g. The preliminary trial consisted of six treatments in triplicate and the densities of U. prolifera were 0, 200, 400, 600, 800, and 1000 mg L−1. According to Khoi and Fotedar (2011) and Wang et al. (2010), to make full use of light and avoid eating by shrimp, U. prolifera were cultured in floating cages which were a strung confinement of PVC rods (30.0 × 25.0 × 20.0 cm) (Fig. 1). The outer walls of the cages were wrapped with multifilament nets with mesh number of 20. Shrimp cultured without U. prolifera were used as the control. During the trial, no water was exchanged. The seawater (salinity 32.5) was disinfected with chlorine bleach, and then, the waste chlorides were removed by deep aeration. Water temperature and pH were 28.5 °C and 8.7, respectively. Dissolved oxygen (DO) remained above 5.0 mg L−1 during the experiment. Light was provided by LED and natural light, and light intensity of water surface remained above 30 μmol photons m−2 s−1 with a 16 h light/8 h dark cycle (Khoi and Fotedar 2011). Shrimp were fed four times a day at a rate of 5% body weight with a commercial feed containing 42% crude protein and 6.5% crude fat (Fuxing, Xiamen) and cultured for 96 h. The daily feed amount was adjusted to the biomass in the tanks and the feeding level was determined at 5% on wet body weight weekly. At 1 h after feeding, uneaten food and feces were siphoned out and filtered through a filter with a mesh number of 200 and then the filtrate water was returned to the same PVC tank. To replace evaporative losses, water was added as needed. Water samples were collected at 24, 48, 72, and 96 h to determine total ammonia nitrogen (TAN), nitrite nitrogen (NO2–N), and nitrate nitrogen (NO3–N) following Ge et al. (2016a).

Fig. 1
figure 1

Plain view of the aquaculture system

Full size image

Main experiment set up

The appropriate range of the stocking density of U. prolifera was maintained considering the water quality (especially ammonia removal efficiency) and the utilization efficiency of light. Considering this, at the main experiment, the stocking density of U. prolifera was 800 g fresh weight m−3 and the density of shrimp was 500 shrimp m−3. The initial weight of shrimp was 3.45 ± 0.13 g. The initial ratio of seaweed biomass to shrimp biomass was 800 g m−3 of fresh seaweed for 1.725 kg shrimp m−3. To test the effects of U. prolifera and the different quantity of water exchange on water quality control and shrimp growth, four treatments were set up in triplicate: 5% (T1), 10% (T2), 15% (T3), and 20% (T4) of exchanged water daily as well as partial water which was exchanged at 1 h after each feeding. The materials and aquaculture facilities were the same as the preliminary experiment. To remove uneaten food and feces from the floating cages and U. prolifera, the floating cages and U. prolifera were removed from the tanks every 3 days, flushed with water, and then weighted to maintain the fresh weight U. prolifera at 800 mg L−1. Then, the floating cages and U. prolifera were replaced in the same tank. The feeding management of shrimp was the same as the preliminary experiment.

Data collection

During the experiment, water temperature, dissolved oxygen (DO), salinity, and pH value were determined daily with an YSI Model Handheld Instrument (YSI Incorporated, USA) at 18:00. Total ammonia nitrogen (TAN), nitrite nitrogen (NO2-N), and nitrate nitrogen (NO3-N) were determined weekly following Ge et al. (2016b). After the rearing test, all shrimp in each tank were counted and weighed and U. prolifera in tank was weighed. The metrics were calculated according to the equations: growth rate (g week−1) = weight gain (g)/culture weeks; cumulative weight (g) = final weight (g)–initial weight (g); survival (%) = [(number of initial shrimp–number of dead shrimp)/number of initial shrimp] × 100; FCR = total dry feed intake (g)/weight gain (g), and specific growth rate (SGR, % day−1) = 100 × ln [final weight/initial weight]/culture time (day).

Statistical analysis

All data are expressed as mean ± standard deviation (SPSS 17.0). A one-way analysis of variance (ANOVA) and Tukey’s tests were applied to determine significant difference among treatments. If significant differences were found, Duncan’s multiple comparison was applied at a 5% significance level.

Results

Effect of U. prolifera on water quality of white shrimp reared in the zero water exchange system

A large number of shrimp in the control group and some U. prolifera in the group of 1000 mg L−1 were found dead in 96 h. During the experiment, the concentration of nitrite nitrogen and nitrate nitrogen was too low to determine, while TAN concentration in all treatments increased over time (Fig. 2). TAN in the control rose sharply from start to finish. At the same time, TAN concentration decreased with the increase of U. prolifera quantity and TAN concentration in the 400-, 600-, and 800-mL L−1 groups was significantly lower than those of the control group all the time (P < 0.05). The concentrations of TAN in the 1000-mg L−1 group from 24 to 72 h were significantly lower than those of the control group (P < 0.05). However, TAN in the 1000-mg L−1 group rose sharply from 72 to 96 h. TAN in the 800 mg-L−1 group that was lower than in the other groups from beginning to end, and at 48 and 72 h, TAN concentrations were significantly lower than those of the 200-, 400-, and 600-mg L−1 groups (P < 0.05).

Fig. 2
figure 2

Changes of TAN concentration in different density of U. prolifera (results presented are mean value and standard deviation, n = 3. Different letters indicate statistical difference at P< 0.05.

Full size image

Effect of U. prolifera and quantity of water exchange on water quality

Water quality during the experiment is shown in Table 1. Salinity, water temperature, pH, and dissolved oxygen (DO) were relatively stable. The concentration of nitrite and nitrate in all treatment remained at very low level from beginning to the end.

Table 1 Water quality in the tanks of four treatments in the polyculture of L. vannamei and U. prolifera during the 35-day experiment
Full size table

TAN in all treatments showed an increasing trend over time (Fig. 3). At the same time point, TAN concentration showed a decreasing trend as the water exchange quantity increased. On days 14 and 21, there were no significant differences in TAN concentration among the treatments (P > 0.05). On day 28, TAN in T4 was not significantly different from that in T3; however, it was significantly lower than in T1 and T2 (P < 0.05). On day 35, the TAN concentrations were 2.75 ± 0.21, 2.47 ± 0.16, 2.22 ± 0.21, and 2.06 ± 0.18 mg L−1, respectively, and the TAN concentrations in T3 and T4 were significantly lower than T1 (P < 0.05); however, no significant differences in TAN concentration were observed in T3 and T4 compared to group T2 (P > 0.05). From beginning to end, no significant differences were observed in TAN concentration between T2 and T3 (P > 0.05).

Fig. 3
figure 3

TAN in the tanks of four treatments in the polyculture of L. vannamei and U. prolifera during the 35-day experiment (results presented are mean and standard deviation n = 3). Different letters indicate statistical differences p< 0.05.

Full size image

Effect of U. prolifera and quantity of water exchange on shrimp growth performance

Shrimp growth performance at the end of the experiment is shown in Table 2. Maximum final weight of shrimp was recorded in group T4, followed by T3, T2, and T1. No significant differences were observed among groups T1, T2, and T3 (P > 0.05); however, they were significantly lower than that of shrimp in T4 (P < 0.05). As is the case with shrimp final weight, maximum growth rate was observed in T4 (1.35 ± 0.07 g) which was significantly faster than those of other groups. There were no significant differences (P > 0.05) in survival between groups T1 and T2; however, survival of T1 was significantly lower than in groups T3 and T4 (P < 0.05). No significant differences were observed among T2, T3, and T4 (P > 0.05). Maximum FCR was recorded in group T1, followed by T2, T3, and T4. No significant differences in FCR were observed in group T2, as compared to T1, T3, or T4 (P > 0.05). Significant differences in water consumption were observed among the treatments (P < 0.05).

Table 2 Shrimp growth performance of four treatments for 35 days
Full size table

U. prolifera growth performance

On day 35, the cumulative weight of U. prolifera showed a decreasing trend as the water exchange quantity increased (Table 3). No significant differences in cumulative weight of U. prolifera were observed among T1, T2, and T3 (P > 0.05); however, they were significantly larger than T4 (P < 0.05). There were no significant differences in SGR and growth rate between T3 and T4 (P > 0.05); however, SGR and growth rate of T4 were significantly slower than those of T1 and T2 (P < 0.05).

Table 3 U. prolifera growth performance of four treatments for 35 days
Full size table

Discussion

Ammonia, nitrite, and nitrate are the main forms of dissolved inorganic nitrogen, and ammonia and nitrite are very poisonous to shrimp (Khoi and Fotedar 2011; Ge et al. 2016a). In the course of production of farming shrimp, ammonia mainly originates from excreta and uneaten feed and ammonia is oxidized to nitrite and nitrate by nitrosobacteria and nitrifying bacteria (Guerdat et al. 2010). According to Kuhn et al. (2010a), there are significant correlation between nitrate and nitrifying bacteria. In the present study, nitrite and nitrate were not detected. This may be a result of the seawater being disinfected with chlorine bleach killing all bacteria (Pintar et al. 2004). It can also indicate that the ammonia nitrogen removal may result from U. prolifera taking up ammonia nitrogen as shown by Baloo et al. (2014). The TAN concentration decreased as the U. prolifera quantity increased, indicating that increasing the quantity of U. prolifera can increase ammonia removal rate. However, TAN in the 1000-mg L−1 group rose sharply from 72 to 96 h. This may be due to some U. prolifera dying because of less or no light, dissolved oxygen and some other conditions. The dead macroalgae cannot remove ammonium and could actually release ammonia. TAN in the 800-mg L−1 group was lower than those of the other group from beginning to end. This indicated that, when the quantity of U. prolifera is 800 mg L−1, the alga has the maximum efficiency of nitrogen removal. However, TAN concentration in all treatments showed an increasing trend over time indicating that Ulva cannot remove nitrogen completely. Therefore, to maintain water quality, some water should be exchanged (Brito et al. 2014).

Temperature, salinity, pH, and dissolved oxygen were within the optimum range for shrimp (Wik et al. 2009; Esparza-Leal et al. 2016) and also within the normal ranges of U. prolifera (Neori et al. 2004; Cruz-Suárez et al. 2010). The concentration of nitrite and nitrate in all treatment remained at very low level throughout. Ammonia is one of the most important factors that impacts the growth and development of shrimp (Vinatea et al. 2010; Ge et al. 2016a). Therefore, ammonia nitrogen removal is a key problem of controlling water quality in shrimp culture (Holl et al. 2011). To remove ammonia nitrogen from shrimp aquaculture ponds, three main practical approaches are emerging: water exchange (Cuzon et al. 2004), bacterial nitrification into nitrate (Kuhn et al. 2010a; Furtado et al. 2015), and seaweed assimilation into biomass (Cahill et al. 2010; Ge et al. 2016b). In this present study, at the same time point, TAN concentration decreased as the water exchange increased. This indicates that water exchange can control water quality. However, water exchange not only increases water consumption (Azim and Little 2008), but also raises the risk of pathogen transmission (Han et al. 2015). Using the seaweed technology for wastewater purification cannot only control water quality (Cahill et al. 2010), but also have economical and social benefits (Cruz-Suárez et al. 2010). Ge et al. (2016a) reported that microalgae co-cultured with shrimp with little water exchange can control ammonia under 0.5 mg L−1. Wang et al. (2007a, b) had similar results showing that Ulva can remove 68% ammonia nitrogen of sea cucumber culture water. In the present study, on day 35, the TAN concentrations were 2.75 ± 0.21, 2.47 ± 0.16, 2.22 ± 0.21, and 2.06 ± 0.18 mg L−1 showing that applying some water exchange in combination with U. prolifera purification can control ammonia nitrogen within safe concentrations for shrimp. From beginning to end, no significant differences were observed in TAN concentration between T2 and T3. This indicated that 10% water exchange and 800 mg L−1U. prolifera can control ammonia nitrogen within safe concentration for white shrimp (500 shrimp m−3).

At the end of the feeding experiment, the shrimp survival rates were between 85.78 and 93.78% indicating that U. prolifera did not harm shrimp. This is similar to the results of Castro et al. (2004) and Cruz-Suárez et al. (2010). Furthermore, Khoi and Fotedar (2011) found that U. lactuca was co-cultured with king prawn that can promote the growth of king prawn. Maximum growth rate of shrimp was recorded in group T4, followed by T3, T2, and T1. This is mainly because better water quality contributes to the growth of shrimp (Copertino et al. 2009) and is consistent with the present result that U. prolifera present in the tank can remove ammonia nitrogen.

In the present study, minimum FCR was 1.48 ± 0.06. The decreased FCR may result from that shrimp cultured with U. prolifera may improve the utilization of artificial feed (Copertino et al. 2009), it may also be because that the seaweed acted as a nutritional supplement for shrimp (Cruz-Suárez et al. 2010). In the present study, shrimp survival rate, final weight, and growth rate showed an increasing trend as the exchange water quantity increased indicating improved water quality (Cuzon et al. 2004; Ge et al. 2016a). In addition, water exchange can promote the sloughing of the cuticle leading to growth (Bray et al. 2006; Kuhn et al. 2010b) thus enhancing feed utilization (Samocha et al. 2015). In the present study, maximum shrimp final weight, growth rate, and survival rate were found in T4; however, they were not significantly greater than those in T3. In addition, water consumption of T4 was 22.42% higher than that of T3. Increasing water exchange not only consumes large amount of water, but also causes pollution of the environment. There were no significant differences in shrimp survival rate, final weight, growth rate, and FCR between T2 and T3 indicating that 10% water exchange and 800 mg L−1U. prolifera can improve the survival rate and growth of shrimp.

Ulva prolifera possesses high nutrient value (Zhao et al. 2011; Teng et al. 2013). Moreover, the macroalgae can be used as an animal feed (Cruz-Suárez et al. 2010) or feed additive (Castro et al. 2004). In the present study, the cumulative weight of U. prolifera decreased as the water exchange quantity increased, probably due to decreased nutrients in the water at higher exchange rates. In the present study, the water temperature was as high as 28.75 °C; however, the seaweed grew fast. This indicated that U. prolifera could adapt to high water temperature and this is consistent with Mantri et al. (2011) who showed that higher temperature enhances U. prolifera growth. Wang et al. (2007a, b) reported that the growth rate of U. pertusa was 3.3%. However, in the present study, the growth rates of U. prolifera were 3.77, 3.80, 3.62, and 3.29%, respectively. This is probably because that the water temperature in the present study was higher than that in Wang et al. (2007a, b) and Gao et al. (2016). No significant differences in growth rate and cumulative weight of U. prolifera were observed among T1, T2, and T3.

Conclusion

This study investigated the effect of integrating U. prolifera with L. vannamei at varying rates of daily water exchange on water quality and shrimp growth. It was found that 10% water exchange daily and 800 mg L−1U. prolifera can control ammonia nitrogen within safe concentration for white shrimp (500 shrimp m−3) and can also improve the survival and growth of shrimp.