BMP Trains

Stormwater Irrigation Volume III: Transport Of The Cyanotoxin Microcystin In Groundwater Beneath Stormwater Ponds: Results Of Soil Column Experiments

Martin Wanielista, University of Central Florida

Abstract

As the demand for fresh water increases in central Florida to meet both public supply and irrigation needs, stormwater is increasingly being managed as a resource to help offset possible future declines in aquifer water levels. Water quality is an important consideration when using stormwater for recharge or harvesting. A constituent of recent concern is cyanobacteria (popularly known as blue-green algae) because some of these can produce toxins (cyanotoxins) that are detrimental to animal and human health. Microcystins are the most commonly found type of cyanotoxin in Florida and have been detected in a variety of rivers, natural lakes, and stormwater ponds. Filtration of stormwater through natural or amended soil sediments and withdrawal of the filtered water through horizontal wells beneath the pond are probable technologies for mitigation of water quality concerns. The potential for transport of microcystin in soil and ground water was investigated by using laboratory soil column experiments.

Results of two soil column experiments indicate the potential for substantial removal of microcystin during saturated flow through sand. Natural stormwater spiked to yield microcystin concentration of 3.9 and 2.2 micrograms per liter was continuously applied during each experiment lasting 3 and 10 days, respectively. Samples were collected from six sampling ports at varying depths along each column in addition to the ponded water at the top of each column. Samples were quantitatively analyzed using enzyme-linked immunosorbent assay (ELISA). Concentration-based microcystin removal efficiencies up to 90 percent and mass-balance based removal efficiencies up to 70 percent were achieved. Breakthrough curves indicated relatively conservative transport of microcystin at breakthrough, suggesting sorption processes were limited, followed by substantial declines in concentration, suggesting kinetic reaction processes were important and possibly caused by microbial degradation. Microcystin losses generally increased with depth up to 2 feet and remained relatively constant at greater depths. Microbial degradation likely can be attributed mostly to the biofilm layer that formed on the sand surface, causing the large microcystin losses in the upper parts of the columns. However, breakthrough curves indicated generally lower concentrations in deeper sampling ports, suggesting additional removal capacity was afforded by the thicker filter bed. Based on the soil and water quality conditions used in the experimental setup, a sand filter bed ranging from 2 to 4 feet in thickness was found to be sufficient for effective microcystin removal.