Effects of detection limits on spatial modeling of water quality in lakes.

Identifying sources and fate of nutrients and pollutants in lake waters is often difficult when key analytes (e.g., dissolved phosphate) are frequently below analytical detection limits (non-detects). One way of dealing with this problem in water quality data is to replace non-detects with "fill-in" values using imputation methods (IMs). While their performance for estimating descriptive statistics (e.g., mean and variance) has been evaluated comprehensively for many environmental variables, whether IMs can reconstruct spatial patterns using long-term water quality data with non-detects under different magnitudes of spatial variation remains under-studied. We developed an integrative framework, combining numerical simulations with univariate and multivariate approaches, to compare performance of nine IMs in recovering spatial patterns of water quality data with different degrees of spatial heterogeneity. We applied this framework to a 12-year water quality dataset sampled from the nearshore region of Lake Ontario near Pickering and Ajax to show the usefulness of IMs in estimating water quality spatial variation. Firstly, in the simplest modeling scenario, we found that most IMs reproduced spatial patterns of univariate data well with ≤30 % non-detects in the dataset. Secondly, when spatial patterns were heterogeneous (e.g., when weak water mixing in nearshore regions limited nutrient transport from input sources to offshore regions), most IMs also performed well by recovering spatial variation in multivariate data with ≤80 % non-detects. Thirdly, when spatial distributions were homogeneous (e.g., when strong water mixing increased transport of nutrients from input sources to other lake areas), only weighted quantile sum regression (WQSR) performed well in reconstructing spatial multivariate data trends with ≤10 % non-detects. Our study highlighted that IMs (especially WQSR) are useful for reconstructing spatial trends of water quality in large lakes. However, potential interactions between spatial heterogeneity and non-detect frequency must be considered when selecting an appropriate IM procedure to accurately model spatial patterns in water quality.

A statistical framework for testing impacts of multiple drivers of surface water quality in nearshore regions of large lakes.

In nearshore regions of large freshwater ecosystems, complex biophysical processes across large geographic regions, combined with the common logistical challenges of data collection by multiple research agencies and shifting monitoring survey designs over time, present challenges for detecting and managing the influence of multiple sources of nutrients and pollution. We present a statistical framework using linear mixed models (LMMs) to test impact of multiple drivers on nearshore water quality of large lakes. Under this framework, we analyzed a 12-year dataset of water quality variables that were measured from a nearshore region along the Canadian shoreline of Lake Ontario (~86 km2), near Pickering and Ajax. Spatial interpolation showed that almost all water quality parameters decreased in magnitude from the shoreline to the offshore. Two exceptions to this nearshore-offshore gradient occurred in a region that extended ~8 km southwest from the outfall of the Duffin Creek Water Pollution Control Plant (DCWPCP) where ammonia + ammonium (NH3+4) increased and pH decreased slightly. Other LMMs combined explanatory factors into major groups (geographic proximities to shoreline tributary mouths and stormwater drains [inflows], tributary discharges, discharge or loading from the DCWPCP, and climate factors). These models showed that geographic proximity to shoreline inflows and/or tributary discharges were the most important drivers for most water quality parameters including concentrations of phosphorus, a key variable for regulating proliferation of harmful algae blooms and nuisance benthic algae in the Great Lakes. Air temperature was correlated with decreased phosphorus concentrations and nitrate + nitrite, whereas total precipitation and snow were correlated with increased concentrations of most nutrients except NH3+4, which was negatively correlated with duration of lake ice cover in winter. Our framework highlights how influence of individual nutrient sources can be distinguished from climate factors within a dominant nearshore-offshore gradient in water quality within nearshore regions of large lakes.