Predicting iron transport in boreal agriculture-dominated catchments under a changing climate.

Increases in iron (Fe) concentration have been reported in boreal regions in recent decades, raising concerns about the fate of ecosystems along water courses. In this study, the SWAT (Soil and Water Assessment Tool) model was applied to the river Mustijoki catchment in southern Finland to determine the current state of Fe transport and to evaluate possible effects of ongoing environmental change in this agriculture-dominated catchment. The model was calibrated using five-year discharge, suspended solids, and Fe data, and validated with a three-year dataset of the same parameters. Further, the model was run with spatially downscaled and bias-corrected climate change scenario data to the year 2100 obtained using five different global climate models. The results were divided into 20-year time steps (2020-2039, 2040-2059, 2060-2079, 2080-2099) and compared against a reference modeling period (1997-2016). With present catchment characteristics of the river Mustijoki, Fe transport was shown to be related to soil erosion and suspended solids transport, driven by hydrological conditions. Arable fields, especially with steeper slopes, were identified as the most likely source of Fe loading. Climate change-induced alterations in riverine Fe transport were simulated as concentrations and as annual mass fluxes. High Fe transport season is already shifting from spring snowmelt events to autumn and winter, and this change is likely to increase in coming decades. Based on modeling results, annual peak concentration in the River Mustijoki was projected to decrease by up to 32% (from 6.2 mg L-1 to 4.2 mg L-1 in scenarios RCP4.5 and RCP8.5) in the coming 20-year period, while lowest winter concentration was projected to increase by 126% (from 1.5 mg L-1 in the reference period (1997-2016) to 3.5 mg L-1 in 2080-2099 in scenario RCP8.5. To compensate for these changes in Fe transport dynamics, water protection and land use management planning must be improved.