中国科学院海洋研究所机构知识库

随着水产养殖业的迅速发展，更多的养殖废水对环境产生重大影响，制约着生态文明建设。循环水养殖系统作为新兴的工业化养殖技术，可减少水资源的用量及实现污水排放管理，在环境上具有可持续。且该系统对养殖环境高度控制，不仅减缓了室外养殖所要承担的风险（包括自然灾害，环境污染，疾病侵袭等），还使养殖品种在全年范围内达到最优生长状况，在经济上具有可持续性。因此，工业化循环水养殖受到越来越多人的青睐。生物滤器作为循环水养殖系统的水处理核心，发展到现在，很难对其功能进行精确预测和控制，因此一直被认为是“黑盒子”。其稳定性与工作效率直接影响循环水养殖的质量与产量。设计适合实际生产的生物滤器是需要解决的首要问题。其次，循环水养殖系统在生产运行过程中水质出现pH值降低的现象，影响生物滤器的工作，导致其水处理能力、微生物群落发生变化。本论文以优化生物滤器运行条件、提高其硝化性能为目标，研究了挂膜启动阶段亚硝酸盐氮积累现象，运行阶段硝化性能和微生物群落对养殖废水pH的响应。本研究得到的结论如下：

1. 在目标产量为7000 kg、养殖密度为35 kg/m³的循环水养殖系统中，日投饲率1%、水温14°C、盐度30‰、pH7.2的条件下，对该循环水养殖系统生物滤器移动床反应器进行设计。结果表明：移动床生物反应器的进水流量为286 m³/h；池子尺寸为4 m×4 m×2.5 m×4个（池中水位2.2 m）；水力停留时间为29 min；换水量为1.16 m³/h；每天的循环次数为20；进水总管、进水支管、补水管管径分别为300 mm、200 mm、20 mm；风机选择为罗茨鼓风JGR150型；鼓风机流量1716 m³/h。该过程对生物滤器的设计思路提出了新的建议，为循环水养殖系统MBBR的完成设计提供了较科学、完整性的思路。

2. 在养殖废水进水总氨氮TAN浓度为2 mg/L、COD浓度为4~5 mg/L、盐度为30，溶解氧为6~7 mg/L、pH为8.0条件下移动床生物反应器从空白填料开始启动构建硝化性能。结果表明：随着生物滤器的启动，氨氧化能力在前36天迅速建立，但是NO₂^--N氧化能力的建立缓慢且出现亚硝酸盐氮积累现象，在第36天达到积累峰值。Shannon、Chao1 和 ACE 多样性指数显示出先降低（T29）后增加趋势，29天后生物膜的功能和稳定性不断增强。AOB的相对丰度和TAN的去除效率成正比，皮尔森等级相关系数为0.624。MBBR启动阶段五个不同的时间(T15，T29，T43，T57，T78)生物膜中AOB与NOB的比值分别是20:1、6:1、4:1、2:1和1:1，该比值变化可以用NOB的定殖速度比AOB较慢来解释。随着生物膜的不断变化，AOB与NOB的比例变得越来越接近1。AOB与NOB的比值变化为解释启动阶段NO₂^--N的变化趋势提供了依据。因此，MBBR启动过程微生物群落组成可动态预测，这为解决启动阶段NO₂^--N积累的问题提供了依据。

3. 在养殖废水进水总氨氮TAN浓度为2 mg/L、COD浓度为4~5 mg/L、盐度为30‰，溶解氧为6~7 mg/L条件下，四组移动床生物反应器在不同pH值进水条件下运行，分别为：7.0（G1）、7.5（G2）、8.0（G3）、8.5（G4）。结果显示：四组反应器TAN转化率均在36天后稳定在80％左右，且没有显著差异，表明pH对TAN的去除没有显著影响。但是表征AOB的OTUs对不同进水pH值较敏感（P<0.05）。研究发现四组实验中存在显著不同的亚硝酸盐氧化性能，其去除效率显示G2> G3 > G1> G4，因此pH值对细菌的影响呈现不对称抑制，即在高pH环境下，pH的抑制作用比在低pH值时更强。PCA、Venn图和方差分析结果显示，不同pH条件下的微生物群落、AOB和NOB之间都存在显著差异。因此，不同进水pH值对于生物滤器的亚硝酸盐氧化性能、微生物群落具有显著影响，且pH值对于细菌的影响呈现不对称抑制，在运行养殖系统时应该时刻监控pH值，在保证安全养殖条件的基础上使pH处于低值，以保障硝化性能及效率较高的微生物群落。

Aquaculture industry is developing rapidly, resulting in increasing volumes of wastewater, which have a significant impact on the environment and prevent the pursuit of ecological civilization. Recirculating aquaculture systems are a new industrial farming technology that fosters environmental sustainability by reducing water consumption and improving sewage discharge management. These systems feature a high degree of control of the aquaculture environment, thereby not only decreasing risks (including natural disasters, environmental pollution, and disease outbreak), but also resulting in optimal growth conditions for the breeding varieties throughout the year while being economically sustainable. As a result, the industrialization of recirculating aquaculture has received increasing attention and support. The core of water treatment processes in recirculating aquaculture systems is represented by biofilters, whose stability and efficiency directly affect the quality and performance of the whole aquaculture system. Designing biofilters that are suitable for aquaculture production and operation is therefore a key step. During the operation of a circulating aquaculture system pH values tend to decrease, thereby affecting the performance of biofilters and their associated microbial communities. To improve biofilter nitrification efficiency, we investigated the effect of different pH values on the nitrification performance of the microbial community growing in biological filters. The main findings of our study are as follows:

1. A designed recirculating aquaculture system had a target yield of 7000 kg and a stocking density of 35 kg/m³, with feeding rate to 1%, water temperature to 14 °C, salinity to 30‰ and pH to 7.2. A moving bed biofilm reactor (MBBR) was specifically designed for our aquaculture system. Four identical biofilters (dimensions: 4 m x 4 m x 2.5 m; water level: 2.2 m) were used to treat wastewater. The influent flow rate was 286 m³/h, the hydraulic retention time was 29 min, the water exchange rate was 1.16 m³/h and the number of cycles per day was 20. The diameters of the total inlet pipe, water inlet branch, and water supplement pipe were 300 mm, 200 mm, and 20 mm, respectively. A JGR150 Roots blower with an air volume flow rate of 1716 m³/h supplied aeration. Our system provided new scientific insight for the design of biofilters and MBBRs applied to recirculating aquaculture systems.

2. The MBBRs were started with blank carriers under the following conditions: ammonia nitrogen concentration of 2 mg/L, COD concentration of 4–5 mg/L, salinity of 30‰, dissolved oxygen concentration of 6–7 mg/L, and pH of 8.0. The results showed that ammonia oxidation capacity developed quickly in the biofilters within the first 36 days of operation. However, nitrite oxidation capacity developed slowly and we observed nitrite accumulation, with a peak on day 36. The Shannon, Chao1 and ACE diversity indices for the biofilter microbial community showed a decreasing trend, while biofilm function and stability increased after day 29. The relative abundance of ammonia-oxidizing bacteria (AOB) was directly proportional to total ammonia nitrogen (TAN) removal efficiency, with a Pearson correlation coefficient of 0.624.

In the early stages of MBBR operations, the ratio of AOB to nitrite-oxidizing bacteria (NOB) at five periods were 20:1, 6:1, 4:1, 2:1, and 1:1, respectively. The observed decrease in the AOB/NOB ratio, which approached a value of 1 as time progressed, was likely due to the slower colonization rates of NOB relative to AOB. The values of the AOB/NOB ratio provided a basis for explaining the trend in nitrite concentration observed during the start-up phase. These results showed that microbial community composition could be predicted dynamically during the MBBR start-up period, thereby providing critical information to address the issue of nitrite accumulation in the start-up stages.

3. Four groups of MBBRs were operated at four different influent pH values: 7.0 (G1), 7.5 (G2), 8.0 (G3), and 8.5 (G4), respectively. After a stable nitrification phase, the ammonia conversion rate of the four MBBR groups reached approximately 80% after 36 days, with no significant difference among groups, indicating that pH had no significant effect on TAN removal. However, the OTUs that characterized the AOB community were sensitive to different influent pH values (p < 0.05). We did find significant differences in nitrite removal performance, with the four treatment groups showing decreasing removal efficiency in the following order: G2 > G3 > G1 > G4. Different pH values showed an asymmetric inhibitory effect on bacteria, with inhibition at high pH values being stronger than at low pH values, as also suggested by the AOB/NOB ratio. Principal component analyses, Venn maps, and variance analyses showed significant differences in microbial community, AOB, and NOB under different pH conditions. Overall, different influent pH values had a significant influence on the nitrite oxidation performance and microbial community of the biofilters, with the effect of pH on bacteria being asymmetrical. The pH values should be closely monitored during system operations and low pH values should be adjusted to ensure safe culture conditions and high nitrification efficiency of the microbial community.