Master of Science
Smart, Chris C
Surface water quality (SWQ) databases have been widely compiled to provide information characterizing environmental conditions. But SWQ databases appear to be under-utilized, given the large investment in their creation. One reason is that database spatial, temporal, and compositional dimension vary through time, reflecting changing priorities through time and contrasts between different agencies, making coherent analysis challenging. This thesis explores the Ontario Provincial Water Quality Monitoring Network (PWQMN) to derive higher order hydrochemical properties, to render SWQ data in “network space” permitting catchment-wide visualization, and in undertaking temporal trend analysis.
Rivers play a critical role in the terrestrial carbon cycle, but the level and role of dissolved carbon dioxide is poorly understood because it is difficult to measure or estimate. A stepwise algorithm was developed to extract an exceptionally large and accurate PCO2 data set from the PWQMN. The results showed ubiquitous supersaturation and decrease downstream, implying high rates of organic matter import into surface waters.
The spatial pattern of surface water monitoring shows a close relationship to a novel upstream ordering system that was exploited to develop a “network space” transformation of rivers and SWQ data. Mapping of chloride, carbon dioxide, oxygen and total phosphorus data in network space showed spatial coherence, clear urban impact, and systematic inter-catchment differences. A complementary mixing algorithm allowed budgeting for high-resolution data sets, but was less successful for general mapping where its value was in auditing the data for point sources or poor monitoring.
Rendering of SWQ data in time using network space was very effective, but risky due to bias and possible errors in the data. Overall, PCO2 levels peaked in the mid-1990s, then fell dramatically to variable, but non-treading levels. These changes were associated with significant transitions in monitoring policy and priorities, so were investigated as possible artifacts. Inter-catchment and epochal differences in PCO2 (and its determinants: alkalinity and pH) were unexpected. This may arise from regional acid rain control programs, but may be a result of contrasting field protocols in different agencies.
Wang, Shuyang, "Development of exploratory data analysis methods for chemical, spatial and temporal analysis of surface water quality data: the Ontario Provincial Water Quality Monitoring Network" (2018). Electronic Thesis and Dissertation Repository. 5905.