The mariculture wastewater used in this study was collected from the drainage channel of the She Pan Tu mariculture garden (29.1538N, 121.5056E) in Ninghai, Zhejiang, China. This 2,141-acre garden encompassed diverse aquaculture practices, including pond aquaculture, aquaculture integrated with photovoltaic power generation, and industrialized aquaculture, thus contributing to the varied composition of the wastewater. Pond aquaculture and photovoltaic-integrated aquaculture primarily cultured Pacific white shrimp (Penaeus vannamei), razor clam (Sinonovacula constricta), blood clam (Tegillarca granosa), and green crab (Scylla serrata), whereas industrialized aquaculture focused on Pacific white shrimp and black tiger prawn (Penaeus monodon). In each experiment, the concentrations of silicate, phosphate, ammonium, nitrite, and nitrate in the mariculture wastewater were measured with various spectrophotometric assays according to National Standards of the PRC (GB/T 12763.4–2007, 2007). The measurement methods included the silico-molybdenum blue method for silicate, phosphor molybdenum blue spectrophotometry for phosphate, hypobromite oxidation for ammonium, diazotization azo method for nitrite, and cadmium-copper reduction method for nitrate. Additionally, the concentration of chlorophyll a was measured using a PHYTO-PAM-II phytoplankton classification fluorometer (Walz GmbH, Effeltrich, Germany).

For this study, four diatom species were selected based on their common use as bait algae in shellfish nurseries. The diatoms Nitzschia closterium, Cyclotella atomus, and Conticribra weissflogii were acquired from Professor Pengfei Cheng at Ningbo University, and Chaetoceros muelleri was obtained from Ninghai Bei Bei Le Aquaculture Company. These species were chosen to assess their growth potential in mariculture wastewater compared to the f/2 medium, a standard culture medium for marine microalgae (Guillard and Ryther, 1962). Initially, mariculture wastewater was collected from the drainage channel, immediately transported to the lab, and sterilized by autoclaving at 121°C for 30 min. The nutrient status of the wastewater was then analyzed (Supplementary Table S1). To ensure consistency, the four diatom species in the logarithmic growth phase were diluted to a same chlorophyll a concentration. Then, 15 mL aliquots of each species were inoculated into both the wastewater and f/2 medium, with each treatment replicated three times. Cultures were incubated in a smart light incubator (GXZ-380A, Ningbo Jiangnan Instrument Factory) at 25 ± 1°C with 4000 lux irradiance under a 12 h:12 h light/dark cycle. Diatom growth was monitored over 7 days by daily chlorophyll a concentration measurements.

This part of the study focused on determining the optimal silicate concentration for the growth of C. muelleri. Five silicate levels (50, 100, 500, 1,000, 4,000 μmol/L) of Na₂SiO₃·9H₂O were evaluated, along with blank and f/2 medium controls, each in triplicate. Mariculture wastewater, collected from the drainage channel, was immediately transported to the lab for autoclaving sterilization at 121°C for 30 min. After assessing its nutritional content (Supplementary Table S1), silicate concentrations were adjusted in the experimental groups. Subsequently, 15 mL of C. muelleri from the logarithmic growth phase was inoculated into the wastewater, mixed thoroughly, and incubated at 25 ± 1°C in a light incubator with 4,000 lux irradiance, following a 12 h:12 h light/dark cycle. Chlorophyll a concentration and silicate content in the culture medium were measured daily and bi-daily, respectively. The experiment was monitored for 7 days.

Mariculture wastewater, collected from the drainage channel, was immediately transported to the lab and sterilized. Its nutritional content was analyzed (Supplementary Table S1). In the first experiment, five concentration gradients (0.2, 0.4, 0.6, 0.8, and 1.0 mg/L) of phosphate were used, as well as blank and f/2 medium controls, with three replicates for each group. In all experimental groups and blank control, the silicate concentration was regulated to 500 μmol/L. In the second experiment, a total of five concentration gradients (0.6, 0.9, 1.2, 1.5, and 1.8 mg/L) were used, along with blank and f/2 medium controls, with three replicates for each group. In all experimental groups and blank control, the ammonium and silicate were regulated to 4 mg/L and 500 μmol/L, respectively. The phosphate concentration in wastewater was adjusted according to the experimental design. Subsequently, 15 mL of C. muelleri from the logarithmic growth phase was inoculated into the wastewater, mixed thoroughly, and incubated at 25 ± 1°C in a light incubator with 4,000 lux irradiance, following a 12 h:12 h light/dark cycle. The concentration of chlorophyll a was measured daily, while the phosphate content of the culture medium was determined every 2 days for 7 days.

Mariculture wastewater, collected from the drainage channel, underwent sterilization for subsequent use. Its nutritional content was analyzed (Supplementary Table S1). NH₄Cl was used as the nitrogen source, with five concentration gradients established at 2, 4, 6, 8, and 10 mg/L, as well as blank control and f/2 medium controls, with three replicates for each group. In the experimental group and blank control, the silicate and phosphate were regulated to 500 μmol/L and 0.6 mg/L, respectively. Subsequently, 15 mL of C. muelleri from the logarithmic growth phase was inoculated into the wastewater, mixed thoroughly, and incubated at 25 ± 1°C in a light incubator with 4,000 lux irradiance, following a 12 h:12 h light/dark cycle. The concentration of chlorophyll a was measured daily, while the ammonium and nitrite concentrations were determined every 2 days over a period of 7 days.

In this experiment, the effects of three nitrogen sources, namely urea, ammonium, and nitrate, on diatom growth were examined. First, the collected mariculture wastewater was sterilized and its nutritional content was determined for subsequent use (Supplementary Table S1). Each nitrogen source was added at a final concentration of 4 mg/L. The experiment included a blank control and an f/2 medium control. All the groups had three replicates for each group. In both the experimental and blank control groups, silicate and phosphate levels were regulated to 500 μmol/L and 0.6 mg/L, respectively. Subsequently, 15 mL of C. muelleri from the logarithmic growth phase was inoculated into the wastewater, mixed thoroughly, and incubated at 25 ± 1°C in a light incubator with 4,000 lux irradiance, following a 12 h:12 h light/dark cycle. The concentration of chlorophyll a was measured daily, while the ammonium, nitrite, and nitrate contents in the medium were measured every 2 days. The experiment was monitored for a duration of 7 days.

The collected mariculture wastewater was sterilized and its nutritional content was determined for subsequent use (Supplementary Table S1). The experimental gradients of trace elements were established at concentrations of 0.5, 1, 2, 4, and 8 mL/L in the wastewater, along with a blank control and an f/2 medium control, with three replicates per group. The 1 L trace element solution comprised 3.15 g FeCl₃·6H₂O, 4.36 g Na₂EDTA·2H₂O, 9.8 mg CuSO₄·5H₂O, 6.3 mg NaMoO₄·2H₂O, 22.0 mg ZnSO₄·7H₂O, 10.0 mg CoCl₂·6H₂O, and 180.0 mg MnCl₂·4H₂O. In both the experimental and blank control groups, the silicate, phosphate, and nitrogen were regulated to 500 μmol/L, 0.6 mg/L, and 4 mg/L, respectively. Trace elements were added according to the experimental design. Subsequently, 15 mL of C. muelleri from the logarithmic growth phase was inoculated into the wastewater, mixed thoroughly, and incubated at 25 ± 1°C in a light incubator with 4,000 lux irradiance, following a 12 h:12 h light/dark cycle. Chlorophyll a concentration was measured daily for 7 days.

The mariculture wastewater, collected from the drainage channel, was initially subjected to different pre-treatments, including autoclave sterilization (121°C, 30 min), UV-C irradiation (254 nm, 20 W, 10 min), and a control group without any treatment. Each group consisted of three replicates. Then the nutritional content of wastewater was analyzed (Supplementary Table S1). Following the pre-treatment, all groups were subjected to the same nutrient regulation strategy with the silicate, phosphate, and ammonium adjusted to 500 μmol/L, 0.6 mg/L, and 4 mg/L, respectively. The mariculture wastewater was placed in the dark for 24 h after pretreatments. Subsequently, 15 mL of C. muelleri from the logarithmic growth phase was inoculated into the wastewater, mixed thoroughly, and incubated at 25 ± 1°C in a light incubator with 4,000 lux irradiance, following a 12 h:12 h light/dark cycle. The chlorophyll a content of diatom and green algae was then examined using the PHYTO-PAM-II phytoplankton classification fluorometer for 7 days.

During the UV irradiation experiment, phytoplankton samples were collected from both raw and 7-day cultivated wastewater using 0.22 μm polyethersulfone (PES) filters (Pall, United States). The phytoplankton DNA on the filters was extracted using the PowerWater DNA Isolation Kit (MoBio Laboratories, Carlsbad, United States). For the microalgae community, the V4 region of the 18S rRNA gene was amplified using the TAReuk454FWD1 and TAReukREV3 primers (Stoeck et al., 2010), following the PCR procedure in our previous study (Zheng et al., 2021). For the bacterial community, the V3-V4 region of the 16S rRNA gene was amplified by the 338F/806R primer pair, with a PCR program consisting of initial denaturation at 95°C for 5 min, followed by 25 cycles of denaturation, annealing, and extension at 95°C for 30 s, 50°C for 30 s, and 72°C for 40 s, respectively, and a final extension at 72°C for 7 min. The PCR products were sequenced on an Illumina NovaSeq 6000 PE 250 platform (Biomarker, China). Quality control and removal of chimeric sequences were conducted using USEARCH v.11 software (Edgar, 2010). A zero-radius operational taxonomic unit (ZOTU) table was generated using the USEARCH denoising algorithm. Representative sequence and taxonomy classification were identified for each ZOTU and aligned against Protist Ribosomal Reference (PR2) database 4.14 (18S rRNA gene) (Guillou et al., 2013) and RDP v18 database (16S rRNA gene) (Cole et al., 2014). The raw sequence data are available in the NCBI database under BioProject accession numbers PRJNA1058243 (18S) and PRJNA1058104 (16S).

The experimental data were analyzed using one-way ANOVA, followed by Duncan’s test for pairwise comparisons using the GraphPad Prism 9. All statistical analyses were conducted with a significance level set at p < 0.05.

In order to assess the adaptability of four diatom species in the mariculture wastewater, the growth characteristics of four diatom species in the wastewater and f/2 medium were compared. The results indicated no significant growth difference among the species in f/2 medium (Figure 1). However, their growth in mariculture wastewater significantly differed from that in f/2 medium (p < 0.05, Figure 1). In the first 3 days, N. closterium showed a marked increase in growth, followed by a decrease. C. atomus exhibited a similar trend, but its growth was slow, with no significant biomass increased after 3 days. C. muelleri and C. weissflogii displayed better adaptability in the mariculture wastewater, with C. muelleri exhibiting a consistent increase in biomass. C. weissflogii initially declined but recovered after 4 days, eventually equalling the final biomass of C. muelleri. Given its robust growth in wastewater, as also supported by previous studies (Karthikeyan et al., 2013; Hemalatha et al., 2014; Saxena et al., 2022), C. muelleri was chosen for subsequent targeted diatom culture experiments. Based on our findings, although the nutrients in the wastewater supported a certain growth capacity for the diatoms, it was insufficient to ensure optimal growth compared to the f/2 medium. Therefore, adjusting the nutrient composition of mariculture wastewater is crucial for optimizing diatom growth.

Silicon is abundant in the earth’s crust and plays a crucial role in diatoms, as they utilize silicon to build their cell walls. Marine diatoms primarily absorb silicon in the form of Si(OH)₄ for cell wall biosynthesis, however the solubility of Si(OH)₄ is relatively low in seawater (<2 mM) (Martin-Jézéquel et al., 2000). Arsad et al. (2019) examined the culture of Haslea ostrearia through different mediums, and found the seawater added with silicate showed better growth. Saxena et al. (2022) found that the addition of inductively coupled plasma nanosilica (ICP-SiO₂) improved diatoms to utilize nutrients from aquaculture wastewater. Hemalatha et al. (2012, 2014) reported that a high chlorophyll a content of Chaetoceros simplex was obtained when the SiO₃²⁻ was 265 μmol/L. In our experiment, we observed a significant improvement in diatom growth with the addition of 500 μmol/L silicate compared to control and lower silicate levels (50 and 100 μmol/L) (Duncan’s test, p < 0.05, Figure 2A). However, further increasing the silicate concentration to 1,000–4,000 μmol/L did not yield additional biomass enhancement (Duncan’s test, p > 0.05, Figure 2B). In all experimental groups, the silicate concentration decreased to below 30 μmol/L after a 7-day culture cycle. This finding is consistent with a previous study, which has shown that 1.5 mM Si (Na₂SiO₃) can be consumed within 5 days by Cyclotella sp. (Ozkan and Rorrer, 2017). These results indicate a high silicate utilization rate and demand in C. muelleri.

Although silicate addition significantly improved biomass in our experiment, the total biomass was still lower than that observed in the f/2 medium control (p < 0.05, Figure 2C). This suggests that there may be other limiting nutrient factors affecting the growth of C. muelleri.

In the first phosphate regulation experiment, diatom growth in all experimental groups plateaued after 3 days, showing no significant difference in the final biomass among these groups (Figure 3A). However, a significant difference was found between the blank control and experimental groups (0.6–1.0 mg/L phosphate) (p < 0.05, Figure 3E), with the final biomass in experimental groups was significantly lower than in the f/2 medium control. The phosphate concentration in all experimental groups was nearly reduced to 0.01 mg/L after 5 days (Figure 3B). It is noteworthy that the ammonium concentration in the mariculture wastewater used in this experiment was relatively low with only 0.3 mg/L (Supplementary Table S1), which may also be a contributing factor to the poor overall growth of C. muelleri.

Therefore, to exclude the potential influence of low nitrogen content on the poor growth of C. muelleri, a second phosphate regulation experiment was conducted. The gradients were adjusted based on the high phosphate utilization rate in the first experiment. Based on the results of silicate regulation experiments, the ammonium and silicate were adjusted to 4 mg/L and 500 μmol/L, respectively. In this experiment, the initial concentration of phosphate is 0.36 mg/L in the control. In contrast to the first experiment, the diatoms in all the experimental groups exhibited robust growth, with higher biomass compared to the f/2 medium control (Figure 3C). However, no significant growth differences were observed between the groups (Figure 3F). The results also revealed that the utilization rate of phosphate by C. muelleri is still very high (Figure 3D). On day 5, the phosphate content of the experimental group with a concentration below 0.9 mg/L was reduced to below 0.01 mg/L, which was similar to the results of the first experiment. Hemalatha et al. (2014) found Chaetoceros simplex obtained maximum growth with the phosphate at 72.4 μM (2.2 mg/L) for 16 days. In the previous study, the dry biomass of C. muelleri with the phosphate at 0.6 and 4.46 mg/L had no significant difference (Lovio-Fragoso et al., 2019). Our results indicated C. muelleri requires limited phosphate, likely around 0.6 mg/L, which can support growth while minimizing environmental phosphorus discharge.

Ammonium is a key nitrogen pollutant in mariculture wastewater (Hargreaves, 1998). Reis Batista et al. (2015) indicated that ammonium nitrogen in the media leads to a longer stationary phase, higher dry weight production, and higher lipid content of C. muelleri compared to nitrate nitrogen. In this study, the silicate and phosphate regulation experiments all indicated that ammonium seems to be critical for the growth of diatoms. Our study examined ammonium nitrogen on diatom growth, finding the diatom biomass in all the experimental groups was higher than in the blank control and the f/2 medium control. The diatom biomass in all experimental groups increased from day 0 to day 5, and exhibited a downward trend from day 6 (Figure 4A). Notably, the diatom biomass increased from the 2 mg/L to 4 mg/L ammonium nitrogen group but decreased from the 4 mg/L to 10 mg/L ammonium nitrogen group on day 5 (Figure 4B). Moreover, ammonium concentrations in 2 mg/L and 4 mg/L groups, as well as the blank and f/2 medium control groups, all dropped below 0.1 mg/L on day 5 (Figure 4C). Additionally, diatoms began to use nitrite nitrogen in the control and 2 mg/L ammonium groups from day 3 to day 7, indicating the shortage of ammonium nitrogen in these groups (Figure 4D). Our results showed ammonium with 10 mg/L had adverse effects on the growth of C. muelleri (p < 0.05, Figure 4B). A previous study also showed that ammonium with 1.18 mM (16.5 mg/L) had toxicity to some algae, but the toxicity disappeared when the concentration decreased to 0.3–0.4 μM (4.2–5.6 mg/L) (Lourenço et al., 2002). Considering the diatom growth patterns and ammonium utilization efficiency, the optimal ammonium concentration for mariculture wastewater appears to be 4 mg/L.

Based on the results of the ammonium regulation experiments, this experiment further compared the effects of different nitrogen sources on the growth of C. muelleri. Diatom abundance generally increased from day 0 to day 5, but showed a decreasing trend since day 6 (Figure 5A). On day 5, the diatom biomass was the highest in the ammonium group, followed by the urea and nitrate groups, and then the f/2 medium group. There was no significant biomass difference between the urea and nitrate groups on day 5 (p < 0.05, Figure 5B). Our study found that C. muelleri with ammonium source had the highest peak in the exponential phase, but the final biomass had no significant difference among the groups with different nitrogen sources in the stationary phases. Liang et al. (2006) found that nitrogen source had no significant effect on the final cell density of C. muelleri at a concentration of 11 mg/L. Reis Batista et al. (2015) also found nitrate source will more easily lead to collapse, and ammonium source has more longer stationary phase and leads to higher biomass finally with a concentration of 16.5 mg/L. Chen et al. (2023) found urea was more favorable than ammonium for the growth of Chaetoceros sp. when nitrogen concentration was 750 mg/L. It is important to note that high concentrations of ammonium have adverse effects on diatom growth. In our study, the ammonium concentration exceeding 10 mg/L significantly inhibited diatom growth. In the previous studies, the ammonium concentration is all above 10 mg/L, which can be a critical factor influencing the results.

During the experiment, the diatoms efficiently utilized both ammonium and nitrate nitrogen, lowering their concentrations in the wastewater (Figures 5C–E). This high efficiency in nitrogen uptake helps explain why the final biomass of diatoms showed no significant differences among the different nitrogen sources. The nitrite concentration initially increased and then decreased in the ammonium and nitrate groups, ultimately being fully utilized by the end of the experiment. The nitrate concentration in the ammonium and urea groups remained at a low level throughout the experiment. We acknowledge that at higher nitrogen concentrations, e.g., 10 mg/L or above, nitrate and urea nitrogen can indeed contribute to higher diatom biomass (Karthikeyan et al., 2013). However, it is not ideal to regulate the nitrate or urea to high concentrations in mariculture wastewater treatment. Manipulating high nitrogen concentrations in the algae pool requires more hydraulic retention time, which is impractical. Overall, our study revealed that ammonium is a beneficial nitrogen source when adjusted to a concentration of 4 mg/L.

In order to investigate the necessity of artificially supplementing trace elements to promote the growth of C. muelleri, an experiment was conducted in which varying concentrations of trace elements were added into mariculture wastewater and the resulting diatom growth was compared. Specifically, the growth trend of C. muelleri and the diatom biomass on day 7 were examined. Our results indicated that the supplementation of trace elements did not improve the diatom growth (Figure 6). Furthermore, there was no significant difference among the experimental groups (Figure 6), suggesting that there was no shortage of trace elements in aquaculture wastewater. It is worth considering that the formulated feed used in aquaculture contains various trace elements (Sarkar et al., 2022). Besides, aquaculture sediment can serve as a source of trace elements and may release trace elements into the overlaying water (Chen et al., 2007). Additionally, the frequent application of fertilizers in aquaculture, primarily composed of nitrogen and phosphorus compounds, may also contribute trace elements to the wastewater (Boyd and Massaut, 1999). Consequently, our findings suggest that supplementing additional trace elements to mariculture wastewater during algae cultivation may be unnecessary.

To develop a practical sterilization technology for diatom targeted cultivation in mariculture wastewater, the effects of UV-C irradiation and autoclave sterilization as pretreatment methods for subsequent diatom growth in wastewater were compared. The results showed that both methods significantly improved diatom growth through the experiment, with the highest diatom abundance observed in the UV-C irradiation group, followed by the autoclave sterilization group, and the least in the control group (p < 0.05, Figures 7A,C). In contrast, green algae was suppressed in the UV-C irradiation and autoclave sterilization groups, but flourished in the control group (p < 0.05, Figures 7B,D). The 18S rRNA gene analysis revealed that UV-C irradiation and autoclave sterilization groups were dominated by Ochrophyta (particularly Chaetoceros), while the control group showed a predominance of Chlorophyta (Schizochlamydella and Picochlorum) (Figures 8A,B). Both UV-C irradiation and autoclave sterilization significantly decrease the relative abundance of Marivita genus (Figure 8D). Interestingly, UV-C irradiation initially increased the relative abundance of Vibrio and Pseudoalteromonas, which subsequently declined after diatom growth (Figure 8D). These findings indicate that the pre-treatment of mariculture wastewater with UV-C irradiation, in combination with the adjustment of nutrients and addition of exogenous diatoms, significantly enhances diatom growth in mariculture wastewater. This method not only lowers microbial diversity but also minimizes the impact on dissolved trace metals, thereby promoting phytoplankton growth (Passero et al., 2014; Chifflet et al., 2019). Moreover, UV-C irradiation did not completely eliminate native Chaetoceros diatom in the wastewater, allowing them to grow alongside the exogenous diatoms after nutrient regulation (Sun et al., 2021). The lower algal and bacterial diversities observed in the treated groups (Figure 8; Supplementary Figure S2) combined with silicate regulation further support the growth advantage for diatoms (Sommer, 1994). However, optimal UV-C irradiation doses and duration for maximizing diatom growth require further investigation. Consequently, UV-C irradiation can replace autoclave sterilization as a pre-treatment method for mariculture wastewater, enabling large-scale wastewater treatment and facilitating diatom targeted culture.