Designing Structures to Remove Phosphorus from Drainage Waters

Waste to Worth: Spreading science and solutions logoWaste to Worth home | More proceedings….


Several groups have developed P removal structures, which are units filled with P sorbing materials and designed to channel runoff water through them while retaining the filter material and P.  The goal is to prevent P from entering a surface water body and allow filtered P to be removed from the watershed after the P-saturated material is removed.  The P sorbing materials utilized are typically by-products from various industries and include steel slag, FGD gypsum, drinking water treatment residuals, and acid mine drainage residuals.  A modeling tool has been developed for (1) sizing a structure based on filter media properties and watershed characteristics, (2) predicting the lifetime of a P removal structure, and (3) estimating total P removal.  In addition to the modeling tool, data from full scale filters will be presented.

Why Are We Concerned About Phosphorus in Water?

Excessive phosphorus (P) in surface waters can result in algae growth, fish kills, eutrophication, and overall poor water quality.  This problem is especially evident in the Illinois River Basin and Chesapeake Bay.  Sources of P to aquatic ecosystems include wastewater treatment plants and also non-point runoff sources (agriculture, horticulture, urban/suburban landscapes).

Soils that have continuously received excess P beyond plant needs typically become “built up” to high levels of soil P.  These soils release dissolved P during rainfall/runoff events.  Current best management practices (BMPs) mostly address particulate P (i.e. P bound to soil particles) transport, not dissolved P.  Dissolved P is more damaging than particulate P because it is immediately 100% available to aquatic life.  Even if all P applications to high P soils are ceased and BMPs are implemented to reduce erosion (i.e. particulate P transport), dissolved P transport will continue to occur for at least 15 years, assuming that plants are harvested from the site.  If plants are not harvested and removed from the site, then dissolved P concentrations may remain elevated in runoff for much longer.

Because soil P levels will remain high for many years, even if P applications cease and efforts are made to “mine” the soil P using plants, the system will continue to “leak” dissolved P during every runoff event.  This has resulted in the need to develop a new BMP that can reduce the transport of dissolved P.

What Did We Do?

Through use of various industrial by-products, we constructed landscape “filters” that remove dissolved P in runoff from “hot spots” before it reaches streams and lakes.  Many industrial by- products that are typically land-filled, including materials produced during drinking water treatment, power generation, and steel production have a beneficial re-use in improving surface water quality by adsorbing P from passing water.

Phosphorus removal structure located at Stillwater Country Club, which utilizes steel slag as the P sorbing material

Several P removal structures were constructed in residential and agricultural watersheds.  Industrial by-products such as flu gas desulfurization gypsum and steel slag were used as P sorbents in the filters.  These filters can be placed in locations known to produce high dissolved P concentrations in runoff; the materials are contained within the structure, which allows them to be removed after they are no longer effective at filtering P (i.e. “saturated” with P).  The materials can then be replaced with fresh materials.  This represents a true removal of P from the system instead of simply tying up the P temporarily.

Automatic samplers and flow meters were used to monitor flow rates and collect samples for measurement of dissolved P and other parameters.  Samples were collected throughout runoff events at both inlet (pre-treatment) and outlet (post-treatment) of structures.  Based on flow volumes and measured P concentrations at the inlet and outlet of structures, P load and P load reductions were calculated.

In addition, approximately 16 different P sorbing materials were tested under flow-through conditions in the laboratory in order to quanitify P removal under different inflow P concentrations and retention times.  With this data set, we have produced a user-friendly model to aid in construction of P removal structures, predict how much P they will remove, and how long they will last until the material is saturated with P.

Phosphorus box filters designed to treat runoff from a poultry farm located in Maryland

What Have We Learned?

Over 8 months of monitoring at the Stillwater site, dissolved P concentrations varied from 0.3 to 1.5 mg L-1 for a residential watershed.  The structure located at that site was able to remove 25% of all the dissolved P that entered the structure over a time period of 8 months.  Other materials can adsorb much more P, but the hydraulic conductivity is much lower, therefore limiting the amount of water that can be treated.  Depending on the material and conditions, P removal structures can pay for themselves if a P trading credit program is ever implemented for non-point source total maximum daily loading (TMDLs).

Future Plans

The computer program model will eventually be made available on the web for practitioners, especially the Natural Resources Conservation Service (NRCS), to aid in P removal structure design.  We are also developing a new P sorbing material that will have greater P sorption capacity and a large hydraulic conductivity, enabling it to both remove P and treat large amounts of water. A large P removal structure is under construction on an eastern Oklahoma poultry farm, where it will capture and treat runoff from areas directly around the poultry houses.

Canister filters placed in a drainage ditchlocated in Maryland


Chad Penn, Associate professor of soil and environmental chemistry, Oklahoma State University,

Josh Payne,  Area animal waste management specialist, Oklahoma State University

Chad Penn, Associate professor of soil and environmental chemistry, Oklahoma State University

Josh McGrath, Associate professor of soil fertility, University of Maryland

Jeff Vitale, Associate professor of agricultural economics, Oklahoma State University

Additional Information

Stoner, D., C.J. Penn, J.M. McGrath, and J.G. Warren.  2012.  Phosphorus removal with by-products in a flow-through setting.  J. Environ. Qual.  41:654-663.

Penn, C.J., J.M. McGrath, E. Rounds, G. Fox, and D. Heeren.  2012.  Trapping phosphorus in runoff with a phosphorus removal structure.  J. Environ. Qual. 41:672-679.

Grubb, K.L., J.M. McGrath, C.J. Penn, and R.B. Bryant. 2012. Effect of land application of phosphorus saturated gypsum on soil phosphorus. Applied and Environmental Soil Science. vol. 2012, Article ID 506951, 7 pages, 2012. doi:10.1155/2012/506951

Grubb, K.L., J.M. McGrath, C.J. Penn, and R.B. Bryant. 2011. Land application of spent gypsum from ditch filters: Phosphorus source or sink? Agricultural Sciences: 2:364-374.

Penn, C.J. and J.M. McGrath.  2011. Predicting phosphorus sorption onto normal and modified slag using a flow-through approach.  J. Wat. Res. Protec. 3:235-244.

Penn, C.J., R.B. Bryant, M.A. Callahan, and J.M. McGrath.  2011. Use of industrial byproducts to sorb and retain phosphorus.  Commun. Soil. Sci. Plant Anal.  42:633-644.

Penn, C.J., R.B. Bryant, P.A. Kleinman, and A. Allen.  2007.  Removing dissolved phosphorus from drainage ditch water with phosphorus sorbing materials.  J. Soil Water Cons.  62:269-276.

Penn, C.J., J.M. McGrath, and R.B. Bryant.  2010.  Ditch drainage management for water quality improvement. In “Agricultural drainage ditches: mitigation wetlands for the 21rst century”.  Ed. M.T. Moore.  151-173.


The authors are grateful to the NRCS, the National Slag Association, The United States Golf Association, and the USDA-ARS for their support of this work.

The authors are solely responsible for the content of these proceedings. The technical information does not necessarily reflect the official position of the sponsoring agencies or institutions represented by planning committee members, and inclusion and distribution herein does not constitute an endorsement of views expressed by the same. Printed materials included herein are not refereed publications. Citations should appear as follows. EXAMPLE: Authors. 2013. Title of presentation. Waste to Worth: Spreading Science and Solutions. Denver, CO. April 1-5, 2013. URL of this page. Accessed on: today’s date.