Development and Applications of a Predictive Model for Geosmin in North Han River, Korea

Abstract

Geosmin is the major off-flavour compound found in rivers and reservoirs during and after cyanobacteria blooms in Korea. A dynamic spatially varying predictive model for geosmin has been developed based on a two-dimensional laterally averaged hydrodynamic and mass transport model, and validated using the field data obtained in North Han River, Korea. The model incorporated the inflow and outflow loadings, production of geosmin by metabolism of cyanobacteria, and the degradation by photolysis, biolysis, and loss by volatilization through a lumped first-order decay rate (K1), which was determined through calibration with the field data. Sensitivity analysis showed that K1 is very sensitive to the simulation results, and should be accurately measured to improve the model performance. The model successfully captured the outbreaks of geosmin production by cyanobacteria (Anabaena spp.) and following decline due to physical and biological processes during the four years of simulation period from 2009 to 2012, except the abnormal winter outbreak occurred in December, 2011. The performance of the model could be enhanced by incorporation of diverse sources of geosmin including benthic source and watershed runoff, and the long-term contribution of cell-bound (particulate) fraction. The model could enable water suppliers to predict taste-and-odor episodes caused by cyanobacteria bloom through simulations of spatial and temporal relations between physical, biological and chemical dynamics in the water body.