2021-2024 Pesticide Risk Metric data with flow and water level

We applied the expanded PRM of Neelamraju et al (2025 - see below) to estimate the combined toxicity of mixtures of 42 pesticides detected in water samples collected by the Great Barrier Reef Catchment Loads Monitoring Program. Mixture toxicity was calculated for every water sample in the time series for all sites and years included in the study, then clipped to the wet season window (first flush + 182 days) for further analysis. Hydrological variables were prepared for the dual purposes of, a) quantification of bias in the existing simple imputation approach, and b) for testing and development of more appropriate infilling methods, with the relative suitability of each variable assessed for this purpose based on model performance and practical applicability. Mixture toxicity estimates for the five PRM groups (other herbicides, fungicides, insecticides, PSII herbicides, and total PAF) were aligned to the flow and time-based explanatory variables using time-matching thresholds: ±30 minutes for discharge data and ±5 minutes for water level data. This process created a matrix of observed and missing PRM data aligned to the explanatory variables, for each of the five PRM groups, for every day of the wet season at each site and sampling year. Neelamraju, Cath A., Mann, Reinier M., Bainbridge, Zoe T., Mitchell, Hannah C., and Turner, Ryan D. R. (2025). Improvement in mixture risk assessment: a multi-national application of the Pesticide Risk Metric. Environmental Advances 22 100676 100676. https://doi.org/10.1016/j.envadv.2025.100676

本研究采用Neelamraju等人（2025年，详见下文）拓展后的农药风险度量（Pesticide Risk Metric, PRM），对大堡礁流域负荷监测计划采集的水样中检出的42种农药混合物的联合毒性进行估算。研究针对纳入分析的所有站点与年份的时间序列内每份水样，均计算了其混合物毒性，随后将数据裁剪至湿季窗口（初冲事件+182天）以开展后续分析。水文变量的制备兼具双重目的：一是量化现有简单插补方法存在的偏差，二是用于测试并开发更适配的缺失值填充方法；针对该用途，我们基于模型性能与实际适用性，对各变量的相对适配性进行了评估。将5类PRM组（其他除草剂、杀菌剂、杀虫剂、光系统II（Photosystem II, PSII）除草剂及总潜在加性因子（PAF））的混合物毒性估算值，与基于流量和时间的解释变量进行时间匹配对齐：流量数据采用±30分钟的时间阈值，水位数据采用±5分钟的时间阈值。该流程为每个PRM组、每个站点每个采样年份湿季的每日数据，构建了与解释变量对齐的含观测值与缺失值的PRM数据集矩阵。参考文献：Neelamraju, Cath A., Mann, Reinier M., Bainbridge, Zoe T., Mitchell, Hannah C., 及Turner, Ryan D. R. (2025). 混合物风险评估的改进：农药风险度量的多国应用. 《环境进展》(Environmental Advances) 第22卷，第100676号文章, https://doi.org/10.1016/j.envadv.2025.100676