Open Collections

Spatial and temporal variation in natural organic matter quantity and quality across a second growth forested drinking water supply area on Vancouver Island, BC

University of British Columbia

Most drinking water in Canada originates in forested headwaters, therefore drinking water security is tied to forest management for many communities. Monitoring source water quality is crucial to a multi-barrier approach to clean drinking water. This research established a stream water sampling program across a second-growth forested drinking water supply area (Greater Victoria, British Columbia, Canada) to evaluate spatiotemporal patterns and variance in natural organic matter (NOM). Over sixteen months (October 2018 to February 2020), 426 stream water samples were collected from twelve sites (ranging from 9.6 to 37 km², elevation 215 to 870 m a.s.l) and analyzed for NOM quality (via UV-Vis absorbance) and quantity (as dissolved organic carbon, DOC). Mean sub-basin DOC concentrations ranged from 4.2 ± 1.8 mg/L⁻¹ to 9.9 ± 3.4 mg/L⁻¹ (DOC range 1.64 - 19.1 mg/L⁻¹). Six of the twelve sites in Leech River watershed (~96 km²) were equipped with vertical passive sampling racks to evaluate rising hydrograph limb NOM dynamics. Hydrologic connectivity to terrestrial source pools increased throughout wet seasons and antecedent wetness was important for stream NOM molecular quality, which shifted from predominantly aliphatic in the dry-season to predominantly aromatic in the wet-season. Sample sub-sets were evaluated for drinking water treatability parameters (n = 8) and total metals contents (n = 42). DOC was correlated with several metals (R² values for Hg: 0.99; Al: 0.81; Fe: 0.72; Cu: 0.47; Ba: 0.25; Mn: 0.21), evidence that aqueous NOM is indeed a master variable important for contaminant transport. More than specific UV absorbance (SUVA₂₅₄) or DOC concentration, the spectral absorbance coefficient at 254 nm (SAC254) was well correlated to disinfection byproduct formation-potential (total trihalomethanes, r = 0.9882; and total haloacetic acids, r = 0.9927). Approximately 80% of the time, DOC concentration peaked with stream stage. Random Forest variable importance measure (RF VIM) identified warm and wet conditions as key drivers for NOM dynamics. RF VIM showed that forest age and harvest history were important predictor variables for NOM aromaticity and molecular size, but that subsurface parent material was a more important driver for NOM quantity and quality.

Text

2020-12-24

Attribution-NonCommercial-NoDerivatives 4.0 International

http://hdl.handle.net/2429/76908

Geological Sciences

University of British Columbia

UBCV

http://creativecommons.org/licenses/by-nc-nd/4.0/