The application of iron-rich sorbents to mitigate phosphorus release from sediments in lowland rivers

Promovendus/a: Lei Xia

Promotor(en): Prof. dr. ir. Erik Smolders

Eutrophication has become a worldwide environmental problem due to excessive discharge of nutrients into water systems. Phosphorus (P) has been identified as one of the key factors responsible for eutrophication. In Flanders, freshwater P concentrations exceed the limits in more than two thirds of water systems, the excesses are mainly pronounced during summer and in smaller rivers or ditches. It has been shown that these summer peaks are neither related to local P emissions nor to agricultural activities, instead it is related to P release from the river bed sediment towards the water. This release is related to anoxic reactions in the sediment, i.e. the reductive dissolution of ferric iron (Fe(III)) (oxy)hydroxides that are strong sorbents for phosphate (PO4). This thesis is devoted to developing and testing remediation techniques to immobilise P in the river sediments. The eutrophication in lakes is often managed by P immobilisation with Fe-rich by-products as adsorbent, however it is unclear if it is also effective in creeks and rivers and it is unclear which properties of the by-products or Fe oxides explain effective immobilisation.

First, static sediment-water incubation experiments were performed to screen effects of dose and properties of the Fe-rich sorbents on P immobilisation. Natural Fe-rich glauconite sand, drinking water treatment residuals and acid mine drainage residuals were added to a river sediment at different doses. The results showed that adding Fe-rich sorbents to sediment can reduce P release from sediment in a dose dependent way, reducing solution P by, on average among all experiments, factor 35 compared to unamended sediments at dose equivalent to about 0.1 kg Fe m2. The P immobilisation efficiency of sorbents was primarily determined by its content of poorly crystalline Fe content. A lower solution pH of the water enhanced P immobilisation but enhanced reductive Fe release to the water which may lower the long-term efficiency in rivers.

Second, experiments in flumes (artificial rivers) were performed with sediments that were either or not treated with glauconite sand. The Fe-rich sorbents amendment effectively reduced the P release from sediment, more so under high flow than under low flow conditions.

A field experiment showed that the high P input to rivers, either from direct emission or from base flow of groundwater, can readily deplete the sorption capacity of the Fe-rich sorbent in rivers. Hence, it is concluded that Fe based P immobilisation is only a viable technology when combined with other management strategies to control P emissions.