Modeling public health effects of climate change: Predicting human exposure to microcystin from Lake Erie

Background/Question/Methods

            Lake Erie is an important economic and cultural resource, but it is threatened by recurring blooms of toxin-producing cyanobacteria.  Lake Erie waters experiencing blooms are regularly found to contain levels of microcystin, a potent liver toxin linked to human illnesses and animal deaths, much in excess of published World Health Organization limits.  These blooms are caused by phosphorous inputs into Lake Erie, the largest of which is runoff carried by the Maumee River.  Climate change is projected to increase the intensity of rain events in this river’s watershed; however, as of yet no model exists which quantitatively predicts the impact that a changing climate will have on public health.

            To fill this gap in knowledge, we are creating a model which will predict human exposure to microcystin given climate and land use scenarios. We first use a runoff model of the Maumee River watershed to predict phosphorous, sediment, and water discharge from the river.  Then, using models published in the literature, we predict the intensity and area of the largest annual bloom on the lake, as well as the associated water-borne concentration of microcystin.  Finally, these are compared to WHO guidelines for microcystin exposure, quantifying the risk to public health.

Results/Conclusions

           Preliminary results with historical data show that the average concentration of Microcystis and microcystin within the peak bloom is relatively stable.  In all years, the concentration of microcystin within the bloom is significantly in excess of WHO guidelines for safe recreational contact.  However, years with more spring rainfall result in a bloom which covers a larger area, meaning that there is more microcystin in the lake.  This means that more microcystin is available for transport to near-shore waters, increasing the risk of exposure.  Over a ten year period (2002 – 2011), bloom area, and therefore total microcystin content, varies as much as a factor of ten.  Future research will attempt to construct a temporally explicit model using monthly data on cyanobacterium abundance in Lake Erie, and data correlating Microcystis and microcystin.  Additionally, fish consumption as an exposure route will be explored.