Effects of landscape variables and season on reference water chemistry of coastal marshes in eastern Georgian Bay

We surveyed 34 marshes in relatively pristine Precambrian Shield catchments in Georgian Bay and related water chemistry to a suite of landscape-level variables, including characteristics of the marsh and its drainage basin. The first landscape principal component (explained 48% of variation) ordered marshes along a gradient with high values corresponding to marshes with large watersheds that contain extensive upstream wetland and that receive relatively high precipitation inputs. This axis was negatively related to specific conductivity, pH, nitrate nitrogen, and SO 4 2– concentrations and positively related to total phosphorus, colour, suspended solids, ammonia nitrogen, and summer dissolved organic carbon. Stepwise regression models built using catchment- and marsh-level variables explained up to 64% of the variation in water chemistry variables. Average precipitation and snowmelt inputs to the catchments were first to enter the majority of models, alone explaining up to 43% of the variation (in the case of water colour), while drainage area alone explained 44% of the variation in pH. Concentrations of catchment-derived constituents in marshes were highest in spring, reflecting greater loadings from the watersheds, while ionic strength was highest during summer, reflecting increased contributions form other sources (i.e., lake water).