Many measurements, tools, and approaches are used to identify and track the influence of human activities on the physicochemical status of streams. Commonly, chemical concentrations are utilized, but in some areas, such as downstream of coal mines, capacity indices such as specific conductivity have also been used to estimate exposure and risk. However, straightforward tools such as conductivity may not identify human influences in areas with saline groundwater inputs, diffuse exposure pathways, and few discharges of industrial wastewater. Researchers have further suggested in conductivity relative to alkalinity may also reveal human influences, but little has been done to evaluate the utility and necessity of this approach. Using data from 16 example sites in the Peace, Athabasca, and Slave Rivers in northern Alberta (but focusing on tributaries in Canada’s oil sands region) available from multiple regional, provincial, and national monitoring programs, we calculated residual conductivity and determined if it could identify the potential influence of human activity on streams in northern Alberta. To account for unequal sampling intervals within the compiled datasets, but also to include multiple covariates, we calculated residual conductivity using the Generalized Estimating Equation (GEE). The Pearson residuals of the GEEs were then plotted over time along with three smoothers (two locally weighted regressions and one General Additive Model) and a linear model to estimate temporal patterns remaining relative to known changes in human activity in the region or adjacent to the study locations. Although there are some inconsistencies in the results and large gaps in the data at some sites, many increases in residual conductivity correspond with known events in northern Alberta, including the potential influence of site preparation at oil sands mines, reductions in particulate emissions, mining, spills, petroleum coke combustion at one oil sands plant, and hydroelectric development in the Peace basin. Some differences in raw conductivity measurements over time were also indicated. Overall, these analyses suggest residual conductivity may identify broad influences of human activity and be a suitable tool for augmenting broad surveillance monitoring of water bodies alongside current approaches. However, some anomalous increases without apparent explanations were also observed suggesting changes in residual conductivity may also be well-suited for prompting additional and more detailed studies or analyses of existing data.

中文翻译：

---

使用加拿大阿尔伯塔省北部河流的历史数据集分析电导率和碱度的关系

许多测量、工具和方法用于识别和跟踪人类活动对河流物理化学状态的影响。通常使用化学浓度，但在某些区域，例如煤矿下游，也使用比电导率等容量指标来估计暴露和风险。然而，诸如电导率之类的简单工具可能无法识别具有咸水地下水输入、扩散暴露路径和少量工业废水排放的地区的人类影响。研究人员进一步提出，相对于碱度的电导率也可能揭示人类的影响，但很少有人评估这种方法的实用性和必要性。使用来自阿萨巴斯卡和平区 16 个示例站点的数据，和阿尔伯塔省北部的奴隶河（但重点关注加拿大油砂地区的支流），我们计算了剩余电导率，并确定它是否可以识别人类活动对阿尔伯塔省北部河流的潜在影响. 为了考虑编译数据集中不相等的采样间隔，但也包括多个协变量，我们使用广义估计方程 (GEE) 计算了剩余电导率。然后随着时间的推移绘制 GEE 的 Pearson 残差以及三个平滑器（两个局部加权回归和一个通用加法模型）和一个线性模型，以估计相对于该地区或研究附近人类活动的已知变化而剩余的时间模式地点。尽管在某些地点的结果存在一些不一致和数据存在较大差距，但残余电导率的许多增加与阿尔伯塔省北部的已知事件相对应，包括油砂矿场地准备的潜在影响、颗粒物排放的减少、采矿、泄漏、一处油砂厂的石油焦燃烧以及和平盆地的水电开发。还指出了原始电导率测量值随时间的一些差异。总体而言，这些分析表明，残余电导率可以识别人类活动的广泛影响，并成为与当前方法一起加强对水体进行广泛监测的合适工具。然而，