Toxicity of sewage effluent to marine invertebrates, tested at Davis 2009/10

Two three-week toxicity tests were completed at Davis station 2009/10 as part of STP project 3217, to provide environmental information in support of an operational infrastructure project to up-grade sewage treatment at Davis (Project 3157). These tests addressed the third specific objective of the STP projects proposal; to determine the toxicity of sewage effluent as the basis for recommendations on the required level of treatment and on what constitutes an adequate or 'safe' dilution factor for dispersal of the effluent discharge to the near shore marine environment. Results from toxicity tests will provide a baseline by which future changes and/or improvements to the effluent discharge can be quantified (in terms of reduced toxicity and impacts to marine biota), if secondary treatment is re-established.The toxicity of the sewage effluent discharge on marine invertebrates local to the Davis coastal area were assessed using standard bioassay protocols developed and tested under AAS #2933 (King). Two key local invertebrate species were used in both tests; the amphipod Paramoera walkeri and the microgastropod Skenella paludionoides. Invertebrates were exposed to varying concentrations of effluent collected from the outfall pipe (test 1) and from a composite made up of samples collected from the main holding tanks around Davis station (test 2). Each test included two controls with 0% effluent (seawater; SW and hyper-saline brine; HSB) and six concentrations of effluent; 3.125%, 6.25%, 12.5%, 25%, 50% and either 68% (test 1) or 63% (test 2). Dilution gradients of sewage effluent were used in tests in order to calculate the dilution of effluent required to cause no observable effects (NOEC), the lowest observable effect (LOEC) and predictable levels of response (e.g. ECX estimates) in test organisms. Organism responses are thereby used to derive protective concentration values for the effluent and the dilution required to have no predicted effects or impact on the community in the receiving marine environment. The behaviour and mortality of the test species were observed on day 1, 2, 4, 7, 10, 14 and 21 of the three week tests. Test 1 (Discharged Effluent):For test 1, effluent was collected directly from the station outfall line at 11am on 15/2/2010. Discharged effluent was collected in a 20 L bucket, placed at the end of the outfall line. Discharge of effluent from the outfall was initiated by pumping from the holding tank of the Sleeping and Medical Quarters / Old Living Quarters by the plumber (communication with the plumber via radio to co-ordinate the time of release from the tank). The effluent was decanted into containers on site and transported back to the laboratory for testing. Test 2 (Composite Effluent): For test 2, a 1 L sample of effluent was collected from each of the Davis station main building holding tanks, using a long handled ladle on 19/2/10. The composite provided an effluent sample that was representative of the waste generated in all activities on station from domestic activities and from work groups. The volumes of each of the building effluents used to make the final composite test effluent were based on estimates of the volume of effluent created in each of the buildings as a proportion of the total stations effluent discharge and were as follows:Building Name; as per map No. 14148,(Building Acronym),volume Summer Accommodation Module,(SAM),416 mL Temporary Accommodation Davis,(TAD),84 mLOperations Building,(OPS),208 mL Sleeping and Medical Quarters / Old Living Quarters,(SMQ/OLQ),834 mLLiving Quarters,(LQ),208 mL Meteorology / Science Buildings,(MET/SCI),84 mLWorkshop,(WORKSHOP),84 mL Climate Processes and Change,(ASP),84 mLLocation of buildings, within the Davis station area are provided in the Davis Buildings and Structures map (Map Catalogue No. 14148), available from the SCAR Map Catalogue at http://data.aad.gov.au/aadc/mapcat/display_map.cfm?map_id=14148. Effluent dilution to test concentrations:Effluent samples in both tests were adjusted to the required test salinity of 32.4 ppt using HSB to match the salinity of the ambient control SW collected off the coast off Davis and used as Control 1 in tests. The HSB salinity adjusted effluent was then diluted with SW to prepare the concentrations series of effluent (as effluent %) to be used as test solutions in toxicity tests. Each test included six concentrations of effluent; 3.125%, 6.25%, 12.5%, 25%, 50% and either 68% (test 1) or 63% (test 2).Controls for both SW and HSB were also required as both were used as diluents in tests. The SW control used seawater from a field aquarium unit* (also the source of SW used to prepare effluent test solutions) and the HSB control was made using milliQ water that was adjusted to the test salinity using HSB. HSB was made by freezing and partially melting control seawater. The melt water is collected as hyper saline brine.*The field aquarium unit contained locally sourced seawater, collected away from known contaminant sources. Water was physically filtered to 1 micron, biofiltered to remove ammonia, UV sterilised and held at a temperature of -0.8 degrees Celsius.Physico-chemical analysis of test effulents:Physico-chemical characteristics of each of the effluent samples including salinity, dissolved, oxygen, temperature and pH were measured immediately on return to the laboratory using a calibrated multi-meter. Further characterisation of effluents, including coliform counts, microbial analysis, organic content and metal concentrations were conducted and are reported in other DAVIS_STP linked data sets. Test species:Two test species were used in both tests; the amphipod Paramoera walkeri and the microgastropod Skenella paludionoides. These test species were chosen based on their abundance and widespread distribution in nearshore environments around Davis Station. Both species were collected by wading and dip netting from the shoreline at Airport Beach, an uncontaminated site located away from the current site of effluent discharge, and for microgastropods, additional sample were obtained from the surface of macro algae collected locally from boats. Both species were housed in control seawater in the field aquarium prior to their use in tests. Collection dates were 8/02/2010 for both species and for microgastropods also 2/02/2010 from Airport Beach (specific dates not known for other sites). Toxicity Test Set up:For each species, each test consisted of 4 replicate 70 mL vials per concentration (including 6 effluent concentrations and 2 controls) containing 60 mL of test solution. 10 individuals were added to each vial at the start of tests (total 320 individuals per species per test). An additional 10 individuals of each species was sampled and preserved in 100% ethanol for genetic analysis, and a further 10 individuals of each test species fixed in 4% formalin for taxonomy and size range analysis prior to the start of each test. Test Conditions and maintenance:Test vials were kept in culture cabinet at 0 degrees Celsius for the duration of the test. In tests with amphipods, a small strip of plastic mesh was added to each vial to provide a substrate and clinging surface for the amphipods. Invertebrates in test vials were fed and average of 0.036 g of Sera granumarin (granulated fish food) on day 6 of each week of the test. Test solutions were renewed in vials on day 7 of each week using freshly collected effluent for test 1 and a new freshly collected composite effluent for test 2.Test Duration: Observations of individuals in test were made at 24, 48, 96 hr, 7, 10, 14 and 21 days for each of the three week tests. This test duration was chosen based on its relevance to the rate of response of both species to effluent exposure. Individuals were scored as either alive or dead, and all dead invertebrates were removed immediately after scoring. A range of sublethal behavioural endpoints including activity were also investigated but did not reveal useful trends and were therefore not used in the final analyses. Test 1 commenced on 16/02/2010 and was terminated on 9/03/2010. Test 2 commenced on 19/02/2010 and was terminated on 12/03/2010.Data Analysis:For each species and each test, and at each of the 7 time end points (24, 48, 96 hr, 7, 10, 14 and 21 days), NOEC and LOEC values were determined using Dunnett's multiple comparison test. Probit Analysis or Trimmed Spearman Karber Tests with Abbott's correction (if assumptions of the Probit Analysis were not met) and Linear Interpolation with Bootstraping were also used to determine point estimates including EC1, 5, 10 and 50 values, and 95 % confidence limits (CL). If ECX, NOEC or LOEC values were outside the range of concentrations tested, results are reported as greater than (gt) or less than (lt) the highest or lowest concentration tested. All statistical analyses were done using the software Toxcalc for Excel (TidePool Scientific Software, California, 1992). To determine the relative sensitivity of the 2 test species and the toxicity of the 2 effluent samples, EC50 values from tests were compared using ANOVA and SNK tests. An average EC50, NOEC and LOEC was determined for the effluent overall. To investigate the precision of the average EC50 estimate and the consistency of responses between tests, the coefficient of variation (CV) and 95 % CL of EC50 values among tests was also calculated.Data Files Provided:Three data files are provided with this record. 1. STP ECOTOX physico_chem data.xlsThis file provides water quality measurements for effluent samples collected in association with the two toxicity tests (Test 1 and Test 2) for the three collection events (day 0, 7 and 14). Parameters measured include: pH, salinity, temperature, oxygen content, immediate oxygen demand (IOD) and FDO (FDO). 2. STPECOTOX Test 1 DISCHARGED.xlsxThis file provides toxicity test data (endpoint observations) for test 1, using effluent discharged from the Sleeping and Medical Quarters / Old Living Quarters outfall. 3. STPECOTOX Test 2 COMPOSITE.xlsxThis file provides toxicity test data (endpoint observations) for test 2, using a composite effluent collected from holding tanks of all Davis station building.

本数据集为STP项目3217的组成部分，于2009至2010年在戴维斯站（Davis Station）完成两项为期三周的毒性试验，旨在为戴维斯站污水处理升级的运营基建项目（项目3157）提供环境支撑数据。本次试验对应STP项目提案的第三项具体目标：确定污水处理出水的毒性，以此为依据提出所需处理水平的建议，并明确向近岸海洋环境排放出水时的合理“安全”稀释比例。若后续恢复二级处理，本次毒性试验结果将作为基准，用于量化评估出水排放的后续变更或改进（如毒性降低、对海洋生物群落的影响改善情况）。

本试验以戴维斯沿海海域本地的海洋无脊椎动物为研究对象，采用AAS #2933（King）制定并验证的标准生物测定方案开展毒性评估。两项试验均选用两种关键本地无脊椎物种：端足类（Amphipod）沃氏拟片钩虾（Paramoera walkeri）与微型腹足类（Microgastropod）斯氏平扁螺（Skenella paludionoides）。

试验1的出水采集自排放管线，试验2的出水为戴维斯站各主储水箱的混合样品。每项试验均设置两组对照组（0%出水：天然海水（Seawater, SW）与高盐卤水（Hyper-Saline Brine, HSB）），并设置6个出水浓度梯度：3.125%、6.25%、12.5%、25%、50%，以及试验1的68%、试验2的63%。通过设置出水稀释梯度，可计算无观察效应浓度（No Observed Effect Concentration, NOEC）、最低观察效应浓度（Lowest Observed Effect Concentration, LOEC），以及试验生物的可预测响应水平（如ECx估算值）。基于生物响应可推导出水的安全保护浓度，以及在接收海洋环境中不会产生预测性效应或群落影响所需的稀释比例。

试验周期为三周，分别在第1、2、4、7、10、14、21天观测试验生物的行为与死亡率。

### 试验1（排放出水）
2010年2月15日11时，研究人员直接从戴维斯站排放管线采集出水：将20L桶置于排放管线末端，通过水暖工无线电协调启动睡眠与医疗区/旧生活区储水箱的泵抽作业，以收集排放出水。出水现场分装后运回实验室开展检测。

### 试验2（混合出水）
2010年2月19日，研究人员使用长柄勺从戴维斯站各主楼储水箱各采集1L出水，混合制成代表全站所有活动（生活与工作产生）废水的混合样品。混合样品各组分体积基于各建筑产水量占全站总排放量的比例估算，具体如下：夏季住宿模块（SAM）：416 mL；临时住宿戴维斯（TAD）：84 mL；作业楼（OPS）：208 mL；睡眠与医疗区/旧生活区（SMQ/OLQ）：834 mL；生活区（LQ）：208 mL；气象/科学楼（MET/SCI）：84 mL；车间（WORKSHOP）：84 mL；气候过程与变化组（ASP）：84 mL。

各建筑在戴维斯站区域的位置可通过《戴维斯站建筑与结构地图》（地图目录编号14148）查询，该地图可从南极研究科学委员会（Scientific Committee on Antarctic Research, SCAR）地图目录获取：http://data.aad.gov.au/aadc/mapcat/display_map.cfm?map_id=14148。

#### 出水稀释与试验溶液配制
两项试验的出水样品均使用HSB调节盐度至32.4 ppt，以匹配戴维斯近岸采集的天然海水盐度（即对照组1的盐度）。随后用天然海水SW稀释调节盐度后的出水，制备系列出水浓度梯度（以出水体积百分比计）作为试验溶液。

两组对照组的配制如下：SW对照组采用野外水族箱的天然海水（同时也是配制试验溶液的SW来源），该水族箱的海水为本地采集且远离已知污染源，经1 μm物理过滤、生物脱氨、UV灭菌后保存在-0.8 ℃；HSB对照组使用密理博超纯水（Milli-Q Water），通过添加HSB调节至试验盐度，其中HSB由天然海水冷冻后部分融解收取高盐卤水制得。

#### 理化分析
返回实验室后立即使用校准后的多参数检测仪测定各出水样品的盐度、溶解氧、温度与pH等理化特征。其余理化表征（如大肠菌群计数、微生物分析、有机质含量与金属浓度）已在其他关联的DAVIS_STP数据集中报告。

#### 试验生物采集与驯养
两种试验生物均采自机场海滩（远离当前排放口的无污染岸线），通过涉水与捞网采集；微型腹足类还可从本地采集的大型藻类表面获取。试验前，所有生物均驯养于野外水族箱的天然海水中。两种生物的采集日期均为2010年2月8日，微型腹足类还可于2010年2月2日从机场海滩采集（其他采样点的具体采集日期不详）。

#### 试验设置
每个物种的每项试验中，每个浓度（含6个出水浓度组与2个对照组）设置4个平行70 mL小瓶，每瓶加入60 mL试验溶液。试验开始时每瓶投放10个个体（每项试验每个物种总计320个个体）。试验开始前，额外采集10个个体用100%乙醇固定用于遗传分析，再采集10个个体用4%甲醛固定用于分类学与体型范围分析。

#### 试验条件与维护
试验全程在0 ℃的培养箱中进行。端足类试验组的每个小瓶需添加一小块塑料网片作为附着基质。试验期间每周第6天，向无脊椎动物投喂平均0.036 g的Sera granumarin颗粒鱼饲料。每7天更换一次试验溶液，试验1使用新采集的排放出水，试验2使用新配制的混合出水。

#### 试验周期与观测
两项试验均为三周，分别在24 h、48 h、96 h、第7、10、14、21天观测个体存活情况。该观测时长基于两种生物对出水暴露的响应速率确定。观测时记录个体存活/死亡状态，死亡个体需在观测后立即移除。此外还考察了包括活动在内的一系列亚致死行为终点，但未发现有效趋势，因此未纳入最终分析。试验1于2010年2月16日启动，2010年3月9日结束；试验2于2010年2月19日启动，2010年3月12日结束。

#### 数据分析
针对每个物种、每项试验的7个时间终点（24 h、48 h、96 h、第7、10、14、21天），采用Dunnett多重比较法确定NOEC与LOEC值。若不满足概率分析（Probit Analysis）的假设条件，则采用Abbott校正的Trimmed Spearman-Karber检验，结合Bootstrap自助法的线性插值法计算包括EC1、EC5、EC10、半有效浓度（Half Effective Concentration, EC50）在内的点估计值及95%置信区间（Confidence Limit, CL）。若ECx、NOEC或LOEC值超出试验浓度范围，则报告为大于（gt）最高试验浓度或小于（lt）最低试验浓度。所有统计分析均使用Excel插件Toxcalc（TidePool Scientific Software，加利福尼亚州，1992年）完成。

为比较两种试验生物的相对敏感性与两种出水样品的毒性，采用方差分析（Analysis of Variance, ANOVA）与Student-Newman-Keuls检验（SNK检验）对比试验得到的EC50值，并计算出水整体的平均EC50、NOEC与LOEC。为考察平均EC50估计值的精度与两次试验响应的一致性，还计算了两次试验间EC50值的变异系数（Coefficient of Variation, CV）与95%置信区间。

#### 提供的数据文件
本记录附带3个数据文件：
1. *STP ECOTOX physico_chem data.xls*：包含两项毒性试验（试验1与试验2）三次采样（第0、7、14天）的出水样品水质检测数据，测定参数包括pH、盐度、温度、溶解氧含量、即时需氧量（IOD）与FDO。
2. *STPECOTOX Test 1 DISCHARGED.xlsx*：包含试验1的毒性试验数据（终点观测值），该试验使用睡眠与医疗区/旧生活区排放口的出水。
3. *STPECOTOX Test 2 COMPOSITE.xlsx*：包含试验2的毒性试验数据（终点观测值），该试验使用从戴维斯站各建筑储水箱采集的混合出水。